Thought Behind Things · Apr 1, 2026 · 1:13:46
Pakistan can't build the LLM. It has to build the application layer.
Dr. Naveed Iftikhar on why the freelancer-sweatshop model is dead, what it actually takes to keep Pakistani founders building from Pakistan, and the lie the local tech ecosystem keeps telling itself about its talent pool.
with Dr. Naveed Iftikhar
12 min read
The episode opens on a day that, by any reasonable measure, should have rearranged how anyone in an emerging market thinks about their career. OpenAI announced it was raising over a hundred billion dollars. Oracle cut thirty thousand people in a single notification, not because the company was bleeding money but because it wanted to redirect eight to ten billion dollars into AI infrastructure. And Anthropic, the company Muzamil calls “the most hottest company” in the market right now, had the source code of its Claude Code agent leak by internal mistake. Three companies. One day. Hundreds of billions of dollars in motion.
Muzamil frames this as the backdrop the rest of the conversation has to sit on top of. The world is shifting to a multipolar order. There is a sovereign debt crisis quietly forming under most major governments. Oil shortages are the first signal, but plastic, aluminium and helium shortages will follow. And then, in his words, “in the backdrop there is an industrial revolution coming.” The question he wants to put to his guest is not whether AI is real. It is whether Pakistan, sitting where it is, can actually do anything useful with the moment.
The guest is Dr. Naveed Iftikhar, co-founder of Atom Camp. PhD in urban affairs and public policy from the University of Delaware. Former governance specialist at the Ministry of Finance. Now training thousands of young Pakistanis and Saudis to work as what he calls AI-first talent. Muzamil introduces him as someone whose background, rural Punjab to bureaucracy to emerging tech, gives him an unusually grounded read on what the technology can actually do for a country like Pakistan, as opposed to “the grifty, oh, AI is gonna change the world” version of the same story.
A bureaucracy that surprised him by saying yes
Muzamil opens with the question that most listeners are probably already asking. Naveed has spent years inside the Ministry of Finance. He is now running AI bootcamps for civil servants. The two worlds, in Muzamil’s framing, feel light-years apart. Pakistani bureaucrats, he says, often look “like they’re living in the nineties.” So is this just bureaucratic theatre, or is something actually happening?
Naveed’s answer pushes back gently on the cynicism without dismissing it. The dominant perception, he agrees, is that the bureaucracy “doesn’t know anything and also they don’t do anything.” But there is also, he insists, a layer of people inside the system who are genuinely eager to learn. Off the back of Pakistan’s recently announced national AI policy, Atom Camp has partnered with the Ministry of IT and Telecom to upskill civil servants, and the surprise, for him, is that the Civil Services Academy, the very institution most people would write off as colonial fossil, was the first to embrace it. AI is now being introduced into common training programmes, into NIPA, into the National School of Public Policy. He is careful not to oversell it: “the scale is still not, you know, wider,” he says. But the door is open in a way he did not expect.
Sovereignty is not just GPUs, it is also your data
The conversation pivots into the harder question. Muzamil lays out the trap with characteristic patience. AI, in his framing, is genuinely a democratiser of intelligence. For a country that has historically lagged in collective intelligence, that is an opportunity. But there is a sovereignty trap: if all the hardware, infrastructure and foundational models sit in the United States or the Gulf, Pakistan ends up “subletting that intelligence from a third party for the rest of our lives.” Every dollar crisis becomes an intelligence crisis.
Naveed agrees, and then complicates the picture. Yes, the democratisation is real. But so is what he calls divergence, “divergence happening among nations, divergence happening among businesses, divergence happening among individuals.” Whoever has more access and more control will pull further ahead of whoever does not. And he insists that sovereignty has two layers, not one. The first is the economic layer Muzamil named: GPUs, cloud, hardware, the dollar cost of compute. The second is data. “Where that data is going? Who is using it? Who is monitoring it?” For a country with Pakistan’s security posture, that is a question that has barely been asked, let alone answered. The Gulf is building hyperscalers. Pakistan, by his read, is at “a very, very, you know, preliminary stage” on both compute and the semiconductor stack underneath it.
The five-layer cake, and the only layer Pakistan can win
Muzamil brings in Faizan Sheikh’s article about the five-point-six trillion rupee question, the piece that argues the old export model is dead and that Pakistan, as a country, has to become a product. Naveed says he read it and agrees. He reaches for Jensen Huang’s recent framing of the AI stack as a five-layer cake, energy, infrastructure, compute, models, applications, and walks through which layers Pakistan can realistically touch.
The honest answer is short. “We cannot be in that race of creating new LLM. We cannot be a new deep seek of the world, phenomena. We cannot create a new open AI.” He is direct about the local theatre around this: “sometimes Pakistanis will go and say the announce, I’ve created new LLM. This is all wrong actually. No one is making any LLM in Pakistan.” What Pakistan can credibly do, he argues, is fine-tune existing models, build small specialised models for narrow tasks, and, above all, win at the application layer.
But the application layer needs a particular kind of talent, and here he separates two things Muzamil and most observers tend to collapse. There is raw talent, the two hundred thousand engineers Pakistan produces every year, which is what Faizan is sceptical of. And there is entrepreneurial talent, which Naveed insists Pakistan does have, and which has been proven globally by the founders who built and exited abroad: the founder of Careem, the recent security exit by Rehan Jalil, the work Khalid is doing on “the internet for the future.” His point is sharp. “You don’t need, like, 500,000 entrepreneurs in the economy. Probably you need, you know, maybe fifty entrepreneurs who will change Pakistani economy.”
The problem is that those fifty currently build in Singapore, Dubai or the United States and keep only back offices in Pakistan. The work, then, is to make it possible to build from inside the country.
What “environment” actually means
Muzamil, characteristically, refuses to let the abstraction sit. “Infrastructure” and “environment” are vague terms, he says. What does it actually take? What is the institutional plumbing? He pre-empts the usual answers, the PayPal complaint he dismisses as “such a, you know, boring, redundant argument” because PayPal is a freelancer tool and the dollar problem runs much deeper than that.
Naveed lists three things, and his list is the most concrete part of the conversation. First: internet infrastructure. Sixty to seventy percent of Pakistan’s geography, by his estimate, has bad internet. He wants the government to look at its own budget honestly, comparing expenditure on motorways with expenditure on internet, and decide what era it is actually trying to build for.
Second: the banking system, and inside that, the way the State Bank treats foreign exchange around startups. He walks through a specific scenario. A founder builds a private limited company in Pakistan. A venture investor puts in one million dollars at a ten million valuation. The company grows to a hundred million. The investor wants to exit at twenty or thirty million. “Now the state bank will come. So many people will come. You know, there is a scam. Where are you taking money? You cannot take.” On paper, he says, everything looks fine. In practice, the exit becomes an interrogation, and “no venture fund will like to come to here.” That single friction, by his read, is the reason Pakistani founders incorporate offshore.
Third: public procurement. In the US, government agencies buy from startups. In Pakistan, contracting rules require billions of rupees in your account and five years of history before anyone in the public sector will touch you. A twenty-two-year-old with a working smart-metering product cannot sell it to a Pakistani electricity company. And once that procurement door is shut at home, the same founder also cannot use a Pakistani reference to go sell the same product into Nigeria or Saudi Arabia the year after. The procurement system, in other words, is quietly killing the export pipeline it claims to want.
Solve the Pakistani problem, then sell it to Africa
Muzamil pulls the thread forward. The old model, he says, was selling a mid-tier service to the United States. The new model is solving a problem in Pakistan, then taking the solution to countries that look like Pakistan: South America, Africa, Central Asia, the Far East. Naveed agrees, and reframes the country itself as an asset. “Pakistan is also a great laboratory for innovation actually because the scale is very big.” A public-sector hospital in Pakistan sees in one day what an American hospital sees in a month. The data, mostly uncaptured, sits there waiting for someone to build on top of it.
Skin in the game, and why he refuses to teach for free
The conversation turns to Atom Camp itself. Muzamil notes, approvingly, that Naveed has resisted the temptation to scale by lowering the bar. There is a barrier to entry. Students pay. Muzamil reads this as a quality signal, and Naveed explains why. There are two dominant models for skills training in Pakistan, and both produce thin results. The first is government-funded free training, which is genuinely free but, on average, “very low quality.” The second is the NGO model, run on donor money. Both, he argues, strip out the one variable that makes adult learners actually show up, skin in the game.
He references Nassim Taleb’s book by name and is unembarrassed about the economics. “If I don’t deliver quality they ask me to give back money.” Atom Camp offers monthly instalments. If a student does not like the first month, they walk and they do not pay for the next two. The structure forces him to keep the bar high. He notes, almost in passing, that the same bootcamp Atom Camp runs for around three hundred dollars sells in the US for six to ten thousand. The point is not the price arbitrage. The point is that even three hundred dollars, in Pakistan, is enough money for a student to take the work seriously.
Muzamil, who has run paid bootcamps himself, jumps in to back this up with a story that lands. He once charged students for a course, showed up to teach four-hour classes, and watched five of forty enrolled students actually attend. “Then, you know, at the end, they’ll be like, oh, you know, the course was really bad. Sir, how many times did you show up?” Free content, he points out, is everywhere, and completion rates on free courses are under one percent. The structure of payment is part of the pedagogy.
The domain knowledge layer, and who actually wins
The most useful career advice in the episode comes when Muzamil asks who the winners will be. Naveed is direct. The biggest loser, he says, is the person who learned only AI engineering or only software engineering with no domain knowledge. The traditional software-engineer-as-coder role is collapsing into a generalist who does the work of business analyst, UX designer, system architect, programmer and tester all at once. “If you are just a coder,” he says, “you are nowhere.”
The winner is the inverse profile. A civil engineer, an economist, an agricultural specialist, a public-health administrator, who layers AI on top of deep domain knowledge. AI gives them scale they did not previously have. And on top of that, he adds a third layer that he thinks is going to matter more, not less, as agents proliferate: ethics, integrity, judgement, the ability to catch a hallucination and decide whether to escalate it. “People are needed, those sort of people who can work with AI but they have their own conscience, ethics, communication and also the domain knowledge.”
Muzamil’s translation of this for the audience is clean. Spend the time building a higher-abstraction understanding of the world, then let AI do the grunt work underneath you. Be the thinking brain.
The lie the ecosystem keeps telling itself
By the end of the conversation, Muzamil asks the question he had been holding back. What lie does the Pakistani tech and startup ecosystem keep telling itself? Off-camera, he says, people are honest. On-camera they tend to give “the sort of rosy, sunny picture.”
Naveed picks two. The first is the talent claim. “We have lot of talent. This is what we say. I say we should convert it into saying we have a lot of raw talent actually.” Raw talent, in his framing, is unfinished material. Talent is what raw talent becomes after it has been shaped, polished and put to work, and Pakistan has been claiming the finished product while only producing the raw material.
The second lie, which he says is the deeper one, is about the universities. He singles out a handful, LUMS, IBA, GIKI, NUST and a few public-sector institutions, that are genuinely producing employable graduates. He notes that the University of Chicago, with whom Atom Camp runs programmes in Saudi Arabia, specifically recruits from LUMS. But across the wider system, by his read, “eighty percent of graduates from Pakistani universities are not employable.” That is the number, he says, that the sector refuses to say out loud.
Muzamil closes by repeating the position that has hardened over the last few episodes of the show. He has stopped waiting for the government. The question, for him, is what the private sector can do now, and how Atom Camp gets from two thousand students a year to twenty thousand and eventually two hundred thousand. The episode ends without a neat answer to that, which is honest. The two things the conversation has made clear are smaller and more useful: Pakistan’s window in AI is the application layer, and whether it gets through that window depends on whether a handful of structural frictions, internet, banking and procurement, get fixed before the rest of the world moves on.
Full transcript
नमस्कार, नमस्तेर, नमस्तेर. Welcome back to the channel. बड़े अरसे से मैं गायब रहा हूं क्योंकि मैं अपने AI के business के ऊपर काम कर रहा था उसकी मैंने कुछ videos भी डाली थी और personally life में कुछ चीज़ें चल रही थी. मैं shift कर रहा था from one city to another. फिर मैं घर ले रहा था and there was a lot of personal things समझें कि updates हैं क्या? सैकड़ों जो news items हर दिन आते हैं उनको हम deconstruct करें और उसमें से देखें कि यार noise क्या है और relevant news क्या है? तो उसको identify करना, उसको deconstruct करना, उसकी relevance निकालना हमारे लिए और फिर साथ ऐसे लोगों से बात करना plan करूंगा वह एक नए model के तहत करूंगा. जो पुराने वाला model था, storytelling वाला model था, long format का model था वह completely change हो रहा है. In fact, अब हम जो है वह ज़्यादा content पर focus कर रहे हैं, production quality पर focus नहीं कर रहे हम हैं. Digital first हो रहे हैं ताकि हमें relevant बंदा timely quickly मिल सके. हम faster shipping to market कर रहे हैं कि सुबह record करेंगे, शाम तक वह upload हो जाएगी. साथ ही मैं ख़ुद personally जो है ना different जो news को मैं unipolar world से दुनिया multipolar world में जा रही है. Iran war is just the tip of the iceberg और उसकी वजह से जो multipolarity है जो यह geopolitical changes हैं इससे सब कुछ append होने वाला है. Already आप देख रहे हैं कि oil की shortages हैं but oil की shortages are just the start. इसकी वजह से plastic की shortages होंगी, aluminum की shortages होंगी, helium की shortages होंगी, and the, you know, consumerist world that we knew will no longer exist आने वाले वक़्त के अंदर and that is going to make major changes in the way that we live life, in the way that we experience life and in way that we do work and business and make money. साथ ही there is an economic reset जो मेरे हिसाब से आने वाला है. There is a sovereign debt crisis पूरी दुनिया की जो governments हैं वह उससे go through कर रही हैं and I think कि वहां पर एक financial reset आएगा, currencies में financial reset आएगा. उसकी वजह से the way that we understand value, the way that we understand finance, the way that we understand investments and growth of our money वह काफ़ी हद तक change होगा. जब भी वह reset आता यह कोई अनोखी बात नहीं है. Multiple times जब super power shift होती हैं, जब status quo shift होता है तो इस तरह के financial resets आते हैं nineteen forty seven में ग़ालीबन. US ने जब last time अपने आप को as a super power emerge कराया था IMF और UN और इस तरह के institutions बने थे that was a financial reset. 1973 के अंदर जब US ने अपने gold को you know अपनी currency के साथ peg तोड़ा था और currency free float essentially कर दी थी gold से हटा दिया था एक नई दुनिया potentially emerge करने वाली है. आज की तारीख़ में अगर कोई कहता है कि वह financial reset के बाद दुनिया कैसी होगी? Nobody knows and that is why it is extremely important कि हम news को continuously देखें. हम देखें कि चीज़ें किस तरह evolve कर रही हैं ताकि हम उसको grasp कर सकें. यह कोई ऐसी नहीं है कि मैं एक दो तीन दिल की बात videos बना लूं और आपको सब समझा दूं. मुझे ख़ुद भी समझना है और आपको भी साथ साथ समझाना है और third जो सबसे important change है मैं multiple times के बारे में बात कर चुका हूं वह है AI जहां पर, you know, या geopolitical changes आ रही हैं, जहां पर economic changes आ रही हैं. वहां पर in the backdrop there is an industrial revolution coming और उसकी वजह से बहुत सारी jobs बदल रही हैं, बहुत सारी changes आ रही हैं and I think it's very very important that we understand those changes and we are ahead of that curve ताकि हम पीछे ना रह जाएं, हम उसकी वजह से jobless ना हो, चाहे हमारे you know जो financial हालात हैं वह potentially disrupt ना हो. तो अगली पांच सौ episodes का मेरा goal बड़ा simple है. मैं foods paradise में नहीं रहना चाहता. मैं अपने आप को एक ostrich की तरह ज़मीन में सिर डालकर बिल्कुल noise को ignore नहीं करना चाहता. I want to understand what is happening in the world and for that reason daily के episodes मैं करूंगा Monday to Friday हर दिन the idea is कि start में कुछ आपसे news share की जाए What I can do, however, is I can make content for the global audience and give them a flavor of Pakistan, give them an understanding of पाकिस्तानी experts and therefore here on forth at least that's the idea. I am going to be creating content in English completely even though that is not completely natural. मेरा जो natural format है वह उर्दू English mix है. लेकिन because potentially कोई गोरा भी देख सकता है, कोई तुर्क भी देख सकता है, कोई Arab भी देख सकता है and आने वाले वक़्त के अंदर चूंकि AI एक ऐसी understand कर सकें ताकि आप उसमें communicate भी कर सकें. Nonetheless, three major things happened in the space of AI today which is what this episode will be covering. OpenAI announced that it will be raising over a $100,000,000,000 which shows how much money is going still even though with the war, with, you know, the fuel shortages, with the fact that a stock market is collapsing and so on and so forth. Still, you know, the AI investment train is going strong and OpenAI is hoping to raise over a $100,000,000,000. God alone knows at what valuation. They're still not going for IPO. They're still going for institutional investors, but there is still a lot of money that is being poured into OpenAI, and that was fascinating to hear. The second more important news that came out was Oracle laying off 30,000 people. A lot of were laid off as well who were working with Oracle all over the world. It was brutal the way that they laid these people off. They literally, you know, people woke up to go to work, they got had a message on their phone telling them that your access has been revoked. You can go to the office, get, you know, collect your personal items. But 30,000 people were laid off with one single notification. And the idea behind it that experts are saying is that they want to free up 8 to $10,000,000,000 that were being spent on these employees to make AI investments. And this is important for you to understand. Oracle, had record profitability last year, they did not have a fiscal problem. They were not going in losses. They did not have a financial reason to lay these people off. It's not like, you know, there was a recession or the economy was bad. They laid these people off because they felt like they could make more money through AI investments, and so they're cutting jobs to invest in AI data centers, AI hardware, and that should show you how serious this transformation is, how serious this revolution is. The number three most interesting news for those who are following AI was Anthropic, which is one of the most talked about, the hottest AI company, even more so than OpenAI right now because of the fact that they have this incredible agent clot code that a lot of companies are using to replace software engineers or software engineers are using to multiply many fold their productivity. The most hottest company, which is actually preparing for a $380,000,000,000 IPO, had their proprietary source code for Claude code leak today. It wasn't because of a hack. It was actually because of a mistake that they made internally, which is why it's out in the open. People are going absolutely nuts. They're trying to sort of reverse engineer it, make their own agents out of it. It's important to understand they haven't released the weights for their LLMs. LLMs are the base layer, the foundational model. And then on top of that, they have a system, a structure, which they use to make their agents work. So the the source code for the agents has been leaked. It is the most efficient, the most transformative agent right now in the market, and, therefore, this leak is going to be used by the Chinese. It's already being used by the open source community to be able to deploy their local agents and use, like, an open source LLM on top of it, thereby reducing the cost and not having to go through Anthropic and having Claude code and, you know, waste tens of thousands of dollars because people who are using this commercially are actually paying for premium subscriptions. They're spending tens of thousands on, you know, tokens. So this leak is going to hit Anthropic a lot. And, obviously, for a lot of people who are in the AI communities, they are very, very excited about this as well. But as you can imagine, this was one day three major companies worth hundreds of billions of dollars moving the markets, you know, major impact to tens of thousands of people. While we're in Pakistan discussing all sorts of random political and, you know, social issues, which, I guess, are fine, But there are major changes happening in the world as you can see from this news. And as the global AI war speeds up, every single country is being forced to reposition itself. And if you aren't building the models, you better figure out how to deploy them, how to, you know, use them in your domains. How do you build products with these models on top and create value. Locally in Pakistan, we recently had the Indus AI week, which for me was very interesting because normally, as a government, as a society, as an industry, we are normally, you know, lagging behind three, four, five years. It was heartening to see that, you know, the industry came together, PSAP was there, PASHA was there, the government was there. And, you know, at the very least, even if we don't have some substantial things coming out of it, we at least had the narrative to be able to look towards AI, and that was great to see. And that is also what brings us to today's guest. He recently announced during the, Indus AI week that he will be equipping Pakistan's bureaucracy with AI education, making them AI first. I found that super interesting. But just to give you a sense of who he is with us today is doctor Navid Iftekhar. He's the cofounder of Atom Camp. I recently mentioned, you know, the boot camp in one of my videos and, you know, a bunch of you guys checked it out as well. He has been training thousands of young talents in Pakistan and in Saudi Arabia to be AI first talents. You know, his students have gone on to work at करीम, Amazon, a bunch of corporate entities in Pakistan and in the region. And he's now expanded his, you know, boot camps and courses to Saudi as well, where there's a huge youth population also interested in becoming AI first. You know that Saudi is doubling down and going massively into AI investments and AI data centers. Doctor नवीद has a PhD in urban affairs and public policy from University of Delaware. He was formerly governance specialist at Pakistan's Ministry of Finance. He has taught economics and public policy at लाम्ज़ and कराची school of business and leadership. He grew up in rural Punjab with an education background, that is what makes him so interesting. He combines these very different areas of rural Punjab and agriculture background to bureaucracy and government and public policy, to emerging technologies and AI, and really has a very grounded understanding of how AI can create long lasting valuable impact. Not the grifty, oh, AI is gonna change the world, cryptocurrency and AI is the future, but a very grounded, very value driven understanding of these technologies and has been on the ground over the last few years working with students, working with stakeholders to, you know, push Pakistan towards the AI revolution. Today, we will be talking about the algebra of upskilling in Pakistan, whether you can actually teach a legacy government machine to survive the AI reset and to understand who wins and loses in the new AI.
वीद्सलाम लिकुम and welcome to the show. बालिकुम अस्सलाम. Thank you, player and honor for me to be here.
I know that you've spent many years inside the Ministry of Finance and now you're pushing AI boot camps. Not something there are two things that are very, very far apart. Particularly, when you look look at and talk to bureaucrats in Pakistan, you almost feel like they're living in a stone age. They're living in the nineties. I'm curious how that transition happened. And more importantly, you recently made an announcement, at the IndusAI week where you said that you're going to be training a lot of these bureaucrats in AI in Pakistan. And I'm curious. I mean, there are lot of headlines that come out of Pakistan regularly. You know, we've been chasing 30,000,000,000, $50,000,000,000 every five years. Nothing sort of comes out of it. We've been trying to transform digitally for many, many years. Whenever the government says something, more often than not, it does not pan out. Let's be real. Is this just bureaucratic theater, or are you actually going to be able to teach a legacy government machine to survive the AI reset? And I'm curious, what are your discussions with the government have been regarding this?
Yeah. So there are two things. You know, first, you mentioned about Ministry of Finance, my economics background. So, you know, many people don't understand how economic growth, modern growth is actually all about technology and innovation. So that's what I studied as part of my PhD at University of Delaware and other things and research that I've been doing. So there is this thing that for growth, for development, then technology and innovation is a very important factor. So, the second thing that you've mentioned, you know, yes, there are challenges in पाकिस्तानी bureaucracy and and there is also diversity as well. The dominant perspective is that they don't know anything and also they don't do anything. But there is also people who are eager to learn, people who are eager to contribute. You know, how much that number is? That's that's obviously a question. But there is a diversity actually. The the the second part that you mentioned about the recently Pakistan has actually announced national AI policy. It has been under discussion. Many people like it. Some say it's a good first step, some don't like it. It's too much wishlist, but nothing is coming out of it. But one thing that came out of it, it was actually upskilling the government workforce in AI. And here is where Atomcam collaborated with Ministry of IT and Telecom, and we've started an initiative to upscale the civil servants in Pakistan. So, yes, there are challenges. Yes, the scale is still not, you know, wider and the bureaucracy that the way you mentioned, the legacy system, the governance systems are all archaic in many ways. But this is just an effort to kick start and it has been quite promising, encouraging particularly. People often think in Pakistan the civil services academies, the colonial legacy, they don't want to change. So, they were the first one actually to embrace it. So, they've finally introduced AI in all of their common training programs and this is where we are helping the governments at every level actually. From civil services training academy to NEPAs, if you are familiar with nationalist, they are, you know, the career progression courses and National School of Public Policy. So, we are introducing AI into all of these training programs. The challenges are there, but I would say that, you know, we've taken a step and there is a huge demand in Pakistan's public sector now for digitalization and for embracing AI.
Right. You mentioned the Ministry of IT. I also know that there is a sort of Pakistan digital authority under Doctor. सुहेल मुनीर that has been created recently. Doctor. Suhail, for people who don't know, was running digital UAE or digital Dubai. So he has a lot of experience in the region. I'm curious, are you guys there are multiple different nodes. I was speaking to someone from PSAP recently. I know there's an incredible leadership there as well. Is there collaboration happening here, everybody's working in their own silos? Because, technically, they're all part of the same, you know, sort of subset. Right? It's digitalization. It's digital transformation. It's AI transformation. How much of collaboration is there is there right now amongst all these stakeholders?
So I can't speak on their behalf. Obviously, that's the government's job to tell how they are collaborating. But I see, you know, that coordination happening and there is a very close working relationship between PDA because I've been part of so many forums and initiatives. So, Pakistan digital authority is actually quite an ambitious mandate that they have And, you know you know, they so so the if you want to implement AI or anything, obviously, you need a digital layer actually. So their main mandate is actually the digitalization. You look at Pakistan, you know, where the scope comes in Pakistan and where the when we see that when you start then the scope is quite bigger as compared to very incremental like in The US economy right now. The health sector of Pakistan is totally need to be digitalized. This is all manual. You know, 99% of it is manual. The education records, the land records, the urban properties, you know, the fintech sector, financial literacy and also, you know, financial penetration in Pakistan is very low. So, this is where the opportunity is. So, I think the PDA, they can speak for themselves, but my understanding is that they are working more on the digital public infrastructure. So that layer which is needed, DPI, on which, you know, you can build AI or other applications. So so that's that's a good initiative. How much they'll succeed? Will it be another authority in Pakistan? That that still is a question, you know.
That makes sense. And for those of you who don't understand, you cannot have AI unless you have the underlying digital data layer that the AI can work on. Right? So so what doctor नवीद is essentially saying is PDA is responsible for digitization of that data layer, particularly in the public domain. I'm curious, though, to to take from the same sort of conversation. What AI is doing right now is that it is democratizing intelligence in a way where the the biggest power or the biggest resource is that said intelligence. Right? Now it can be an incredible opportunity because we are an sort of illiterate nation. And the reason why we lag behind more often is that we do not have that sort of intelligent minds, the the the overall collective intelligence that a society or an economy needs. AI solves that. And then using that intelligence, we can solve problems, create value, so on and so forth. But I think there is a sovereignty problem there as well where who owns that intelligence? Do we end up sort of subletting that intelligence from a third party for the rest of our lives? There's a lot of money today being spent on the hardware and infrastructure layer. And the reason for that is because, again, if that hardware and infrastructure layer is in The Middle East or in The US, it's a dollar problem again for us. And every time there is a, you know, current account deficit or a boom bust, we will be sort of, you know, we won't be able to access that intelligence anymore. So I'm curious, is AI still a buzzword? Or or are people in the government, people around you, are they able to understand, sort of break it down, what this revolution truly means? And how do we incrementally, in a structured way, make sure that we can fully utilize its potential as a sovereign, sort of developing nation, if that makes sense?
Yeah. I think that's an important question. And I would say that realization is there, but efforts are still very nascent, actually. So, you know, realization like you mentioned about equalizer or democratization of AI. At the same time, there is also divergence happening. Divergence happening among nations, divergence happening among businesses, divergence happening among individuals as well. And divergence is that if you have more access to AI, you have more control on AI, then you are more advanced than the other one on the spectrum. So that, you know, there is this opportunity as well and there is this risk as well. So it really depends how nations, individuals, businesses embrace, adopt and get something out of it or they lose out of it. There will be gainers. There will be, not everyone is going to gain. So, sovereignty is a big big issue. As you know, there are two part to it. You know, one is is what you are talking about, the economic security that we are talking about, the dollar problem that Pakistan has buying cloud or other products, farm applied or GPUs or, you know, hardware. But the second problem is also where your data security lies in. So, you know, Pakistan has its own, you know, security paradigm and all sort of challenges, political, social, security related. So, how that data is going to add a new challenge for that paradigm as well. So, where that data is going? Who is using it? Who is monitoring it? So there is this realization, but the challenge is huge actually for a country like Pakistan. You know, like in The Middle East, they are investing huge money in data centers. They've they've hyperscalers, what we call them AWS, NVIDIA and with all partners they are working. Pakistan does not have that much capability. We are at a very, there are data centers. Pakistan government is supporting, entrepreneurs are working on it. But it's a very very, you know, preliminary stage of data centers and very low compute available at the moment. If you want to have more compute, you need more investment. And for sovereignty, this is just one layer buying GPUs or cloud. The second layer is actually the real part is where the Silicon Valley takes advantage what we call the semiconductor, the chip market. So even if you can buy, if but you cannot produce tomorrow, you know, you don't have access to those chips or, you know, then you are nowhere in AI. So so, Pakistan is also started this initiative of semiconductor in Pakistan. They are working on it. You know, obviously, that's a very very preliminary stage and it's not going to bring results sooner. But realization is there, what you asked. How much effort is happening? Maybe at a very initial stage. But the opportunity is there for investors, for entrepreneurs, for venture funds to come and, you know, invest in such areas.
Makes sense. There was a very interesting article by Faizans Dike, the R5.6 trillion rupee question recently that came out on the news. He basically speaks about how, you know, we we make these sort of tall claims of $30,000,000,000 of exports. The government has been saying that continuously. But he sort of talks about how the old model is dead. You cannot have, sweatshop, like a BPO, low value sort of freelancer economy that's gonna take you there. You actually need to build products. But more so, what he argues is you have to Pakistan or Pakistani industry has to be the product itself. I'm curious now. In a world where there is sort of this fight between US and China and there are two these two sort of opposite powers within the AI arms race as well. Separately, there's obviously a lot of geopolitics happening. But within AI, you know, we're seeing the the deep seek versus the open AIs of the world. We're seeing NVIDIA not being able to sell to China, China making its own indigenous chips, and, you know, this constant this battle of who is going to come out as the superpower of AI. In that context, where do you see Pakistan fitting in? And considering, you know, you've been able to you've been training talent in Pakistan and in Saudi, and that talent, in my understanding, has been working on global prod projects, has been earning in dollars, has been been able to monetize that skill. Right? So as a as a macroeconomy, as a as a developing country, I don't think we can follow the Indian model as they've done over the last thirty years. It has to be indigenous. I do believe it's an incredible opportunity, and it's sort of the great equalizer where we don't have to compete. We have to have we potentially can have the first mover's advantage. But what do you see Pakistan's role in this sort of, weird, aggressive, multipolar dimension.
Yeah. So I agree with Fazan article. Actually, I read it, and it said it's a very good observation. And this is what I also advocate that the traditional model is not going to work. So Pakistan, like, earlier we were talking about all these layers in AI, like, you know, recently popularized by Jensen Huang and of NVIDIA. And, you know, all these five layers from energy to infrastructure to compute and all that, you know, that's an interesting analogy that he created a five layer cake actually recently. So so, you know, there is Pakistan does not have so much energy. Pakistan does not have so much compute. What does Pakistan has? You know, we cannot be in that race of creating new LLM. We cannot be a new deep seek of the world, phenomena. We cannot create a new open AI that we should accept it, that's fine. So, day sometimes पाकिस्तानीs will go and say the announce, I've created new LLM. This is all wrong actually. No one is making any LLM in Pakistan. So, they're all fine tuning LLMs and that's fine. You know, technology is there, fine tune LLMs, make SLMs, small language models very specialized for particular tasks. Maybe that's possible for Pakistan at some stage. So, Pakistan right now where is advantage of Pakistan is actually talent. तो talent is not what we think that's the raw talent. When we talk about 2,000, you know, 200,000 engineers producing and that much number that is raw talent. That what फैज़ान is saying. But there is another mic, you know, there is another opportunity that Pakistan has. Actually, पाकिस्तानी talent is very entrepreneurial and it is proven in the world. You know, you talk about मद्दस्तर शेखा of cream, how he built it. It's a पाकिस्तानी. Recent very good exit by a पाकिस्तानी. I'm forgetting his name रेहान or something security. That's a recent exit happened in विकान Valley. You follow राहेब हुसैन, you follow रेहान जलील. हां, इसे रेहान जलील security. You know, follow there is another guy खालिद who is building the internet for the future. So, you know, पाकिस्तानी has, you don't need like, you know, 500,000 entrepreneurs in the economy. Probably you need, you know, maybe 50 entrepreneurs who will change पाकिस्तानी economy. So, those 50, Pakistan has that potential actually. So right now these पाकिस्तानीs go abroad and build there and exit there. So what is for Pakistan if these people can build in Pakistan in future and the application layer actually. So we can be at application layer. We cannot compete in data centers. We cannot compete in LLMs. So there is this potential that we need to create. And for that, we need to have better environment for entrepreneurs to build in Pakistan so that they don't go and open a company in Singapore or Dubai or you know, they go US, they go everywhere but they at least they have an environment that they can build while while sitting in Pakistan. Right now, these talented people that I mentioned, they have back offices in Pakistan.
Right. You've mentioned providing the sort of infrastructure or the environment. These are vague terms. Right? I'd I'd I'd be curious if you could expand on that. What would it take? What is the institutional plumbing, for example, that is required? Because we've heard this before as well. I remember there was this sort of software technology. I don't even remember. The last government had some they had some something of their own where they were like, oh, you know, special technology zones or something like that. The new government has with SIFCs and the, you know, PDAs and so on and so forth. They keep on making these organizations. But, unfortunately, I feel like nobody really understands. We talk about PayPal not being in Pakistan. I mean, I find that to be such a, you know, boring, redundant argument. You don't do enterprise work on PayPal. PayPal is great for freelancers, but if you wanna do billions of dollars of, you know, business, it's PayPal is not the problem. It's just, you know, it's a gisty bitty argument that keeps on popping up. I'm assuming, and and I know this for a fact, dollar is a problem. Your your finance, you know, your overall financial economy where dollar flow in, flow out has always been a problem, and therefore, there there is always holdcoast. The only the bottom of the barrel cost comes to Pakistan. The money stays in Dubai or Singapore or so on. But have you done some sort of a homework or have an understanding of what these sort of structural changes are that might lead to people staying in Pakistan and building in Pakistan? And I'm not even going to I mean, I'm going to ignore completely the the sort of security and law and order and political challenges, and I'm gonna exclusively focus on the business side of things.
So I'll I'll mention three things, actually. Let me if I come to the fourth one as well. So, you know, I'm not talking about organizations and infrastructure like typically पाकिस्तानी is thing creating a new organization, creating a new park and, you know, I'm all for parks but that's not that's that's not something that I'm talking about when I say environment. I'll tell you a few things only. Number one thing is the is your Internet infrastructure. The speed, the connectivity, the accessibility, still, you know, about I would say that about geographically, the 60 to 70% of पाकिस्तानी geography, internet quality is very bad. You know, other than the big cities, you know, Internet quality is not good. You need more bandwidth. You need more reliable Internet. This is this is and I always say that to the government, and and you mentioned about my finance background, and I look at budgets as well, and I raise this question. How much and again it's a गिस्या पिता argument that motorways versus this thing and this thing but I'm not saying they are mutually exclusive that if you build internet you don't build motorways but I still say compare Pakistan's expenditures on internet infrastructure versus on roads actually. Actually. That should tell you that in the modern era this infrastructure is very important. So, is the first thing. The second thing is पाकिस्तानी forget PayPal and everything is पाकिस्तानी banking system actually. So, you know, fintech has has revolutionized. You know, you you you can do some data and you can tell people that how how inaccessible banking was in Pakistan and now with all these digital wallets some movements at retail level has already started. But you know still पाकिस्तानी banks will not extend credit cards they don't take risk actually. So, the digital money is still very limited penetration in Pakistan. You know, e commerce has to happen. It happens on cash on delivery. You know, because of people don't have these all digital wallets, credit cards. So, this whole infrastructure of fintech and modernizing Pakistan's banking system is the second layer that we need to work on. And within that layer, you know, there is this there is this thing that they treat all these IT and the way they do controls on forex in Pakistan. Like, let's say, I'm I'm giving you one example. Today, a startup raises money. And, you know, in Pakistan, sitting in Pakistan, they have a private limited company in Pakistan and there is money. And today this valuation is $10,000,000 and they, you know, there is, let's say, 10%. The $1,000,000 comes in Pakistan and today the company becomes $11,000,000. And company raises a product within a year, you know, company gets another money and they say that we are $100,000,000 company, you let's say the new valuation and they do an exit further, you know, they do scale a, you know, round and, you know, series a round and they get another $4,050,000,000 or whatever. At some stage, the venture investor who came from abroad, they want to exit. And now they are taking their money, you know, maybe, you know, like, 20,000,000 or $30,000,000. And now the state bank will come. So many people will come. You know, there is a scam. Where are you taking money? You cannot take. You cannot do this. On papers, they will say everything is fine. But when you will start taking your money abroad, there will be hundreds of questions on you. And no, no venture fund will like to come to here. This is the second thing that I want to mention. The third and the last thing that Pakistan has actually, the real problem is the public procurement in Pakistan. So, I always give example like in The US and all these people, the government agencies, they buy from startups, they buy from all these innovative companies. In Pakistan, you have traditional system of contracting. You must have been doing for the last five years, you know. You must have billions of rupees in your account. And now you are a kid out of a university. You have built a brilliant application to, for smart metering for an electricity company. And no electricity company will never buy from you, you know. So, this public procurement system because once you, I always give example, once you your your startup sells this product to a पाकिस्तानी electricity company, tomorrow they will go and sell it in Nigeria. You know, the next day they will go and sell in Saudi Arabia. So, this public procurement system, so I want to highlight these three things which are very important for Pakistan is I will again repeat it the internet infrastructure, the financial system and the government regulations, relating to public procurement and taxation. So I think these are the things for improvements.
100%. I think one one very interesting point that you mentioned towards the end was पाकिस्तानी is solving problems in Pakistan and being able to solve that, then the same problem in Africa. A lot of people, the old model that people are thinking about is how do we go and sell a service like a mid tier, low value, low cost service to The US? And Faizan's article mentions the same thing. When you're talking about productizing your industry and and solving problems as innovators, you will be going to South America. You'll be you will be going to Africa. You will be going to a lot of countries that are fairly similar to you. The model of business may not be as it used to be, know, $5,000 per month. I'll work like a slave and, you know, do brain dead work, but rather going in and solving a problem for them that can create value. And I think that's a shift that a lot of people need to really begin to think and and begin to have, like, that pivot towards Africa and Central Asia and and, you know, even Far East and South America as well. In terms of I wanna channel your sort of boot camps and your AI education because this is something that's been very close to my heart. I've mentioned this multiple times on the channel. I'll keep on repeating it till people in Pakistan take it seriously. We have the second largest population of outdoor school children in the world. We, you know, we have every year 5,000,000 kids turn the age of 18. Half of them do not graduate high school. Out of the 2,500,000 that do, only one fifth or 500,000 end up going to a university. Now in the age of AI, you actually genuinely need higher value understanding of the world. The way that I'll I'll break it down not as degrees or or or skills or so on and so forth, What you need to do is you need to spend time in having a a a higher understanding, on a on a higher abstract layer, of the world, and then utilize that to be able to solve problems and leverage AI for that matter to be able to do the grunt work so that you can be the thinking brain. Right? Unfortunately, as a as a country, we do not have the we're not spending the resources, to train that talent. And, again, I don't want to have, like, a soft story about how the government is crappy and they're not spending enough. I'd be curious to figure out, what the private sector can do. And so I'm curious. You've been teaching a a bunch of these kids. You haven't gone out with, like, this scalable crazy model of teaching million kids. Right? Because I feel like that is unsustainable, and that reduces the quality. You've been very focused, and you've really worked on a granular level with these kids. I want to understand if you can break down the exact variables, that stop from scaling in Pakistan, whether that's a language barrier, that is brain drain, systemic friction, is it that there is not enough teachers available? What stops you and generally the overall education sector from scaling and sort of allowing for the 2,000,000 high school graduates every year that can't pursue higher education to be able to do so?
Yeah, I I think there are so many questions. The first one you mentioned about, you know, I I, you know, that the way you framed it. And I've also mentioned that Pakistan is solving पाकिस्तानी problems and then going global. So, Pakistan is also a great laboratory for innovation actually because the scale is very big. So, you know, you name any sector and that's what, you know, even if someone is not पाकिस्तानी and they want to work in Pakistan, so this is, this offers them scale. You go to a public sector hospital and, you know, and and then you go to a hospital in The US, you know. And just imagine, in an emergency, maybe in a day there will be a few 100 people walking in with some pain, you know, in a US hospital. And then you go to an emergency in पाकिस्तानी hospital and you will see that hundreds of people in one day walking on with different challenges, different and OPDs if you say, maybe thousands, you know. So just imagine the data that we have that is uncaptured. And anyone working on education numbers you've told about how many kids we have, what we can train them, what is the impact, we can track them. And this can bring, you know, that sort of opportunity for anyone. So so in Pakistan, what we've been doing at ATOM camp is actually we focus more on the white collar jobs and recently we've started on grey collar jobs as well, not on blue collar. Does not mean which are not important but this is our specialty. And the scale is that in Pakistan there has been, you know, these slogans to train so many people. It's not possible to do at scale for private sector. So, sector can only do, you know, limited interventions. The scale is with the government, you know, where, you know, maybe 200,000 people are graduating every year from universities. And now if the government improves the university education then the scale will come. I can train like maybe two to 3,000 people, 2,000 max maybe in a year And, that is how, the scale and and not me, you know, if there are, you know, hundreds of us doing similar things and there are some doing. So, we'll remain at that level. So, what we've been doing in Pakistan, we thought that Pakistan, there is a gap what we are, you know, learning at universities and what the market And, that gap we are trying to fill with our boot camps, whether those are AI boot camps, data boot camps. And what we try to do, we take a bunch of people, we train them live, we do professional development. And then I feel very happy if someone's sitting in चित्राल because most of our boot camps happen online. Someone sitting in चित्राल does their boot camp and they succeed in entering into, you know, United Nations job. They succeed into entering into corporate jobs, international jobs, Middle East, US. So so what we try to solve the problem is is the fundamental gap what you are saying that universities are not realizing. I'll not go into that detail. University education is still important. We also take people who are enrolled in universities, who are graduate of universities because still it offers that exposure. But over the time, Pakistan scaled a lot university education without improving the quality actually. So, what the private sector faces no hindrance in many ways is the mindset that, you know, the people have. There are less challenges from the government side for the private sector, more challenges in पाकिस्तानी youth who still think they'll get a degree and they'll get a job. You know, that's what they need, you know. Second thing, I will go abroad somewhere and I will get a job. You know, the international markets have become very competitive. पाकिस्तानी markets have become very competitive. All what I advocate, you know, pick up any skill. And AI is obviously a top layer that any domain knowledge you do have, and that's what we advocate to people. Then they come to us, they learn AI. I tell them, get the domain knowledge. You know, you need to have agriculture knowledge to apply AI in agri tech. You need to have health sector knowledge to apply AI in health sector. You need to have education knowledge to apply in in in, you know. So, what we train people in our boot camps, not just, you know, AI algorithms, all these things, but the domain knowledge, professional development, how to apply AI. So what we've succeeded so far, I always tell people there are lums, IPA, nursed, you know, maybe three and GIK four, five universities and people come from those universities to our boot camps as well. But in Pakistan, my goal is usually a student of University of मरदान can get a job equivalent to a student of, you know, lumps. You know, if not equivalent, subpar, maybe closer to it. And in career, they can reach there as well on average. So so this is what our goal is at our boot camp. So I'll stop here. But the main thing is that how we can, you know, the second year university, third year university kid, and we can bring it, you know, somewhere where the first year university kid is studying and, you know, working.
That makes sense. Whenever you're talking about AI, and I've really thought deeply about this because there is a lot of grifting happening in the overall globally when it comes to AI. There's a lot of fear mongering. And whenever there is fear mongering, there will be people who will come and sell you fear and take money. When it comes to courses and boot camps, for example, people tend to automatically assume that you're out here, and you're trying to just sort of take money from them, and it's not going to lead to any outcome, particularly in Pakistan because universities have failed our students as well, and so it's understandable if they are cynical. The reason for mentioning this, by the way, for the audience, you know, I've seen the work that doctor Navid has done. I've said it in a previous video as well. He doesn't just allow anyone to come in because, technically, he can. There's so so much fear in the market. If he opens it right now, he can make millions overnight. And whether he delivers or not, that's another question. Right? But he's focused. He takes very specific limited people. There are barriers to entry. I feel like the way to actually look for quality education is if there is a barrier to entry. If they're taking everyone, you know, if you're a if you're a pass and I will make you a PhD scholar, you know, it's not gonna work out. But, if there is some sort of a barrier to entry more often than not, that's the right one. But I'm I'm curious, in a wider conversation, how do you sort of differentiate between the AI grift where people are like, oh, you know, this tool came out, and it's going it's killing Hollywood. And this this new tool came out, and it's going to change everything. It's not. There's a logical transition period. And maybe change is happening. It's real. It's serious. But there is a question mark about real value creation as well. Right? So how do you sort of differentiate between the the sort of grifty, chuna AI versus high quality, utilitarian, money making, value addition AI?
There are there are you know what? The problem that we solved with AtomCamp, and that was very unique when I started. Many people stopped me doing it. They said no one will come to you. People will not come without any international credibility and, you know, people will not pay. So, there are two problems in Pakistan that I have. I am coming from public sector, development sector and there is two problem. One, you know, sometimes not at the scale of international level but the government also funds lot of skill trainings in through so many institutions and those trainings are free for people. You know, they are happening in AI, they are happening in everyone. So, the quality of those training programs is often very low. You know, some people also graduate from there with their effort they can also but on average the quality is very low. So it's free but it's low, very low quality and people don't end up going anywhere. The second dominant problem has been in Pakistan the NGO model that you go to donors and you'd go every day and ask them to give you money and then you train people. And again, you do it free. So if you have money, you do it free. That's totally okay. But I'm I'm not a businessman that much wealthy person. So I'm just, you know, what I wanted to do, I created, wanted to create a sustainable model. People pay the fee, not like, you know, the boot camp, like, I just want to give you an example. In US, the same boot camp that we offer, it's it's like $300 or so. In US the same boot camp is being sold from $6,000 to $10,000 and and even more we give more credit hours, more training But even I understand for पाकिस्तानीs paying $300 is also not a small amount. But this is just to meet our expenditures. What we problem solved today I take 30,000 per month from someone and if I don't deliver quality they ask me to give back money, you know. So and it happens sometimes out of like 50 people, maybe one or two don't like our boot camp and they ask us to give our money back and we give some time, you know. We give two weeks actually. So, you study, you say and you take money back if you don't like it. We have monthly installment. You paid one month. You didn't like the next two months. Don't pay us. And now for me to continue, I need to offer that quality, you know. If I'll offer that quality, they'll pay for the next three months. Otherwise, they'll go away. So I wanted to create a model where, you know, the you've you've read that book, नसीम निकालेस्तालेव is is a famous book that skin in the game, you know. So so, you know, if I'm not charging, then people come and, you know, the government can do it. You know, government should do it free. The government can do it, but for private entrepreneur. So this is the problem we try to solve, people pay us and now coming to that point where you say that fear mongering, yes this is happening. I also talk about that AI will replace many jobs but AI is not going to tomorrow eliminate everything that people say and the new world is going to AI has its own limitations, it will take some time. Still the domain knowledge will matter a lot with AI, not just AI. So, this is the area if you are a mechanical engineer, electrical engineer, you are electronics person, you are agriculture, you are sociologist, you are a booed economist and now use AI. So, that is the winning game. So I tell to people that you learn a tool and now you'll win the world that's wrong. So any degree that you are doing, you can learn AI. That's true. But how? So, like, our boot camp, like, the three months AI boot camp is only for those people who are tech backgrounds, who have computer science, engineering, who have studied mathematics, who have studied statistics, programming, some programming exposure. They can come directly to three months. Ok. So, you've not studied all these fields. You are actually, social sciences background, economics, political science, humanities, English literature. Still you are a university undergrad student or you have done an undergrad or exceptionally if you are a good student you can be an intermediate or you know O level A levels. So you can do a three months data analytics boot camp with us where we build the foundation. We build the logic, statistics, we teach SQL, Python. We teach Power BI, Advanced Excel, programming, EDA. And if they pass that three months, they can also enter into three months AI. So then we build the foundation first. But the tools trainings like we have some boot camps automation with AI. Up now there are the tools training. The tool training is also not for a you know five year kid can come for fun and they can do all these things but we don't enroll them. We don't deal much with that sort of audience. So we deal with the workforce level. So we advocate it to people who are already in the field. They are HR manager, they are procurement manager, they are general manager manufacturing, they are sales officer. They join these boot camps. They learn automation tools and they apply them that. So this is the right way to navigate AI, not just jumping on one tool and mastering it and saying I've learned AI. So this is this is our philosophy at AtomCamp.
That makes sense. And personally, if I if I've I've dived little bit into education and training and and so on, again, something that's a bit of a passion for me. You mentioned skin in the game, you mentioned it for yourself. Right? I feel like even for the students, most people don't realize it. They feel like they want to have free content available. The fact is free content is available all around us. It's it's available on Coursera. It's available on 10 different places. The completion rate for for free courses is more often than not less than 1%. When students do end up paying money, they have the skin in the game as well, so they tend to show up. A teacher cannot teach you if you don't want to learn. And so I feel like I am personally opposed to the idea of free education. I think charge them 500 rupees. If if, you know, that's, even if the government has to do it, just charge them something based on their own sort of, you know, economic status. But they have to have some money going out where they realize, okay. If I'm spending money and if I don't show up, I'm losing something. Because I did a couple of boot camps. I even charged people money, and you'd be surprised. Like, two months, I would be like, I'd be like, I took money from people. I have to show up. I I would do, like, four hour classes. And if there are 40 people who would be show up or who had signed up, three or four or five would show up, and there'd be a list of excuses. Oh, you know, this happened and that happened. I was like, I'm here putting my time and energy and effort to make sure that you can learn something, I'm happy to teach you whatever, however I'll come to your level. But, you know, then people don't show up. And then, you know, at the end, they'll be like, oh, you know, the the course was really bad. Sir, how many times did you show up? How much how many times did you even try? So I feel like the skin in the game aspect is is very much required from the student side as well. But sort of changing the conversation a little bit, You did mention, you know, you mentioned that the government should be spending on this. You mentioned that, obviously, the Ministry of IT and and and you guys are collaborating to teach the bureaucrats AI as well. How do you justify? You've been in the Ministry of Finance as well. You know where Pakistan is right now. The debts are high. The current account deficit is a forever problem. We don't have dollars. We don't have money. How do you mathematically justify allocating capital to tech innovation right now? For the vast majority of people in Pakistan, they're right now thinking, what is AI? Why do I need AI? Right? For them, the bigger question is, why can't we solve the food security problem? Why can't we solve the energy security problem? And particularly with the Iran war, I feel like a lot of these more fundamental problems are going to take center stage as shortages grow and, you know, there is instability in the region. So how do you, at this very, very interesting junction, sort of justify allocating capital to tech innovation and AI?
You know, when I would put it in a way that tech innovation and AI not just for gold to look fancy or too cool that we are doing AI. You've mentioned so many problems which are going to emerge after Iran war and, you know, so many things. So let's see if if, a, can AI, our tech help us to solve that problem with a lower cost the way we've been doing it? Then, yes, embrace AI. So not, you know, imposing AI. We want to do AI in any way. You know? Yet that that's what, you know, that obviously we need to pursue what the world is doing, see around and see things. But let's say we want to solve mobility problems because in Pakistan, you know, you see that, you know, we have not very good public transport. So, we spend lot of money on our cars. How to build a public transport system that can serve everyone? So, this is a challenge. Can tech and AI help us to solve that problem? So if yes, then we should adopt it. You know, agriculture, you know, food security is going to become problem for many parts of the world. Pakistan has some problem, but still we are blessed with fertile lands and waters to some extent and, you know, everything. So, we have challenges in food security too, but still agriculture is is still what we can invest in. But, you know, for a farmer, how they can better take care of their animal with AI? How they can, you know, there is a recent Indian startup. You may have heard about that the caller that they've done with this with the cause and, you know, they've raised money. So, this whole market is available in Pakistan. So, you know,
so I get you upraising the call. You mentioned this. I just wanna I just I don't know if you if you remember Adnan who made Kowler in Pakistan. It's so fascinating. The the current startup that raised money. There was a Pakistani startup, Kowler. Ten years ago, they were building this, And it's shocking to me that today, it is raising billions of dollars. Like, it's at a billions of dollars of valuation. Again, it comes to the point of we did not have the infrastructure available for our startup to be able to sort of do this on a global scale, unfortunately.
Yeah, that's what I'm saying. You know, this is where, and what you were mentioning, I would just disagree with that. The government should not do free. Government should do free, but free should not be low quality, you know, one. Second, the government there should be for higher education, I still believe university education should not be for everyone. So, I believe in more filtering for if government is paying. If you are paying from your pocket and you don't deserve a university education, your your merit, you know, does not qualify, then you pay for it. But the government should pay to talented people free. That's totally fine. But now the education has changed actually. What we need to understand. So, earlier the model was you do a degree, you go to the work and stay there for your life in that corporate sector and and on the same job and the same hierarchy. But now the model has changed actually. The degree that you have entered into, it's not enough. And even if university does a lot of effort, the student needs to make some effort to learn what is happening in the world, what skills are happening. So, don't shift all the blame to the university. The the individual need to take charge now. And like you mentioned, if people don't want to join our boot camp because we charge, there are so much free things happening. So, are people who learn free. All what we do at boot camps, we create discipline, mentoring, coaching, organization, curation that you will probably take a year in going to one course and other course and other course and the roadmap that we've created it will take you there in three months. We charge for that. But if some there are so many people who learn themselves they are disciplined they can go they can learn free. So student need to consider that even during degree, after degree, continuously you've joined a job, you are in HR now, you are in procurement, you are in manufacturing, you are in sales, in marketing, You still need to do courses, you still need to do. तो every time government will not come to you, you know what you need to learn. There are governments which are doing their limited population growth like in Germany and other places. In our case, there's a huge population growth. Continuously government funding those people for life are the choices spending on the primary quality education. You know, where schooling education can get better and better are there. So there are choices that the public money we have to make. So I think that the government need to invest in basic education, very good high school, totally free in Pakistan. That is wrong the way we spend, you know, so many money on private schools. That is where the government need to spend. The moment you enter into university, selected very few universities, very talented and then private sector if they want to teach people go there and go there. So for the private sector, the market is actually I think is the skills market. We are AtomCamp, there are hundreds of AtomCamp are needed, somewhere specializing in marketing, someone specializing in agriculture skills, in farming, in plumbing, in refrigeration. This whole market, the apprenticeship system like the way Germany has, the similar system I'm these days trying to work with Pakistan. There is a largest network of पाकिस्तानी technical and vocational training colleges and schools. So, I'm I'm trying to work with that network in टैफ्टा, in Punjab and other places where they have basic skills like, you know, I don't know if you know about these diplomas in electrical, electronics, three year diploma that they do after matriculation. No, what I'm asking them, teach them electronics, teach them mechanical, teach them civil engineering diploma, but include AI tools relevant to that civil engineering. So, is where we are coming in as a private sector, advising them, helping them and also training that workforce actually. But the government need to invest more and more money on the basic education infrastructure.
Makes sense. In terms of the jobs, now I'm hearing a lot. I mean, I've I was roaming around in The US, speaking to a lot of tech leaders. And and what I'm hearing, again, in The US is that it's the worst time for fresh graduates, and I'm seeing that as well. We we we saw Oracle lay off 30,000 people today. Meta has all major tech companies are laying off. What they're saying is people who already have experience, they can sort of multiply their productivity using AI. But if you are a fresh entrant, you're you know, nobody wants to invest time in you. More often than not, the previous model was fresh graduates were inducted into companies. They weren't as productive or valuable for companies, but they were they companies would invest on them because they would be their leadership over time. Right? That was the rotational model. Unfortunately, companies are no longer getting that fresh talent because they can get that bottom of the barrel low value work being done by AI, and they want more experienced resources. In terms of and, again, that for Pakistan is is a scary, you know, thought because and not just for Pakistan, to be very honest. I mean, there's lot of chatter about India losing out on billions of dollars of of IT services exports because a lot of that sort of medium and low value work is going to be automated, courtesy of the cloud code and so so on and so forth. Who do you think will be winners in all of this? For, again, for younger people who are looking to start their careers, who are trying to figure out how do we what do we learn? How do we plug ourselves into this sort of new world that's emerging for the youth in emerging markets in the next five years? What are areas that you think are actually capturing value? And what do you advise?
Yeah, so there is there is couple of like the layoffs and obviously the the the shift in Pakistan will also happen but it has still not as a context in Pakistan the wages, the social security is still unfortunately or fortunately that's very low. So still people can hire a young graduate with the less cost than a cloud premium, you know, of $100 a month. So, that's something that still happens in Pakistan. So, still new talent is being hired in Pakistan. That's what I'm telling people, you know, everyone who was thinking go going abroad. So, international market is now more tougher than पाकिस्तानी market. So so on that side, you know, I don't know. No one knows actually. Everyone is predicting, but still we need to be clear that the future is very very uncertain right now that we don't know much about it. But but everyone can make intuition based on their what I understand. If you are learning just AI engineering, you are learning software engineering, you are doing IT degree and you don't have any domain knowledge. You are the bigger loser than, than a civil engineering engineer actually. So, that is what in Pakistan or in the world the traditional model of software engineering has changed. Software engineer was meant to be a coder to programmer. Someone will go to the client. There will other person, the forward would know these days we call forward engineer. Earlier we call to use the client facing, the sales. The business analyst will come. The UI UX person will come and they will design it. The system architect will come and then they will give you, you will code and program and you know then someone else will test and you know that. And now all of these things they're asking one person to do it, you know, and and launching the shipping the product. So, no Yeah. If you are just a coder, you are just traditionally they were software engineers but they were forced into become coders actually. Not everyone but majority of them. So that model will change. Now if even if you are a software engineer I tell people study health sector you know what is happening in health sector? How health management information system works? How you know hospitals work? So how the polio vaccination happens, how the you know this vaccination system works. So, all I am saying that if you have even a technical background like a software engineer, AI engineer but you don't have domain knowledge then you will lose. The second person that you will lose, you have civil engineering knowledge, economics knowledge. AI now offers the scale, to you that you can do much bigger, more broader and impact ful work with AI that you used to do before. So now the loser is the one who just learned AI and did not learn the domain knowledge, and also the loser who learned the domain knowledge. And and then the third layer in this is actually now going to be very important is actually your professional skills, your interpersonal skills, your integrity, your ethics. These things are going to become more important now as compared to, just the mechanical task because mechanical task will happen in any way. People will look for who are ethically good people. If AI agent is deployed, you are supervising and you've seen some hallucination, you've seen some, you know, escalation where you need to escalate, whether you have escalated or you ignored it. You know, people are needed, those sort of people who can work with AI but they have their own conscience, ethics, communication and also the domain knowledge. So, this is where the right approach is that I advocate to पाकिस्तानी youth to everywhere. We've recently moved. I earlier I've mentioned Pakistan how is a good laboratory. We did this model and we expanded to Saudi Arabia and that was also an opportunity for us and we are doing the similar thing there. You have a domain knowledge and now you can build with AI further and enhance your scale.
Makes sense, Doctor. Navit. We're at the one hour mark. I'm going to wrap this up. I just want to know one thing, I feel like, I mean, a lot of these conversations that I've had, people tend to give the sort of rosy, sunny picture. Right? Off camera, more often than not, we tend to then talk about real problems. I wanna I want you to channel your sort of cynical energy and give me one lie that the local tech and startup ecosystem in Pakistan tells itself. Because we we I feel like we share a lot of lies. Right? We we say, oh, you know, we have this incredible youth population. We have incredible amounts of talent, so on and so forth. The fact is we have not been able to convert that into actionable value. And so I'm curious. What frustrates you when you go out and you hear people regurgitating some lie? What do we need to be honest about ourselves with?
I just want to say we have we have lot of talent. This is what we say. I say we should convert it into saying we have a lot of raw talent actually. That is ready to be defined and ready to be polished. So, we have raw talent, not the talent actually. So, this is the lie that we have. And when I look at our quality, the biggest लाई is our university, the quality of our university education. There are few universities which are doing amazing job in Pakistan and people from there like, I work with universities in The US. We've collaborated with some of the universities. I just want to tell you like with University of Chicago we are doing a few programs together at in Saudi Arabia and they tell me how many students they recruit and they prefer from lums from Pakistan. So, there are these institutes in Pakistan where I also taught there and so, you know, there are such institutes Lums, IBA, GKE, NURST, they are providing quality education and some public sector. But the difference in public sector is, like, I often tell, like, from lumps, there is a graduation class of computer science and economics. 70% will go graduate education abroad or they will get good jobs in Pakistan, you know. 70% you can say and 30% may not be able to do better. But in these public sector universities, second tier universities, probably 10 to 15% will be doing great, 80% are not doing great. So, this is where Pakistan needs to work actually. Those 80% graduating, 20% still with their effort, with their, you know, motivation, they are doing better. But this is where the lie is there. I think that 80% graduates from पाकिस्तानी universities are not employable. So, this is this is the lie that we have and we need to fix it actually.
Makes sense. Doctor नावीद, this was incredible. Thank you so much for taking the time out. I'm excited to see how you guys grow further from here. I would be curious to know how you can scale this because to be very honest, I mean, I know you're still hopeful and you're like, you know, government needs to do this and government needs to do that. I've made a made a mantra. I try to ignore everything from the government. And I say, okay. What can we do now as a private sector to be able to solve these problems? I feel like the moment the private sector can come in and start solving, things will be fine. But, obviously, it's great to see that optimism, and I'm curious to see how you can take that number from 2,000 to 20,000 to 200,000 eventually with partnerships. Thank you so much and for for the consumers, for the viewers. If you guys are interested in the sort of work that doctor नवीद is doing, check out AtomCamp. We're gonna put a link down in the description. This is not a sponsored podcast at all. None of the podcasts coming on and coming in on this show will be sponsored. These are genuinely good people doing genuinely good work. I am handpicking them. I'm looking around. I'm finding, you know, people who are doing incredible work. More often than not, the people who are doing great work tend not to sell themselves. They sell themselves short because they're so focused on the work. And so I thought it would be useful to utilize my platform to give them some exposure as well. Sir, thank you so much for taking the time out. Thank you. Thank you. Play here. And for all of you guys, thank you so much for watching. Let me know what are your thoughts in the comment section below. Are you scared of AI? Are you upscaling yourselves? What domain are you participating? Like, what domain are you from? And how are you applying AI in your domains? Would love to know your thoughts in the comment section below. But nonetheless, was the game of Avaya Sensei for watching Thought Behind Things. Thank you so much for watching, and I'll see you in the next one.
More from Thought Behind Things
Jun 30, 2026
Simon Dixon: the US empire is being wound down by design
with Simon Dixon
Simon Dixon argues the American empire isn't collapsing — transnational capital is deliberately unwinding it, and Pakistan's military has become its leverage in the transition.
Listen →
52:45 Jun 9, 2026
How Asad Mehmood landed Mattermost from Pakistan before A levels
with Asad Mehmood
Asad Mehmood walked into Mattermost before he had A levels, crossed two million dollars on Upwork, and now runs a design agency from Pakistan. He sat with Muzamil to lay out the framework underneath it: become undeniably good, then become visible, then sell outcomes.
Listen →
59:04 Jun 5, 2026
A 22-year-old Pakistani's road map to space by 2050
with Hassan Mossin
Hassan Mossin, a 22-year-old Pakistani astrophysicist training with the International Institute for Astronautical Sciences, walks Muzamil through the next twenty-five years of space — commercial stations by 2030, a lunar base by 2032, and why the bottleneck isn't food or oxygen, it's energy.
Listen →Never miss what's next.
The dispatch - new writing and conversations, straight to your inbox.
First name, last name, email - in your inbox weekly. No spam.