The Machine that Changed the World, vAI

Transcribed by qwen3-asr-1.7b with a schedule prompt. Possible prompt artifacts, names heard more often here than in the unprompted Voice Memos transcript (Qwen vs Voice Memos): Tariq Shaukat (1 vs 0), Sonar (3 vs 2), Evan Kormos (1 vs 0). Check them against the audio before quoting.

00:00:06 — Toyota's Quality Principles in Software Manufacturing

Morning, everybody. Hi, my name is Tariq Shaukat, and this is Sonar, as you just heard, and we're here to talk today about how we have been helping companies build software factories that actually work for enterprises, and I thought it'd be an interesting way to start, to start looking at actual factories because we know how to build actual factories, and for those of you who have a little bit of the gray hair and all of this sort of thing, you may remember the 1980s.

And in the 1980s, you could find cars that were really good cars, but if you look at the picture here, and if you remember those times, the cards that didn't quite fit together perfectly. Right, you'd have different pieces that would leave big gaps. It would be really good cars that just didn't kind of work the way they were supposed to, and then Toyota and particularly Lexus came out with a commercial, and it showed something very simple, which was a ball bearing running across the seams of this car, and it was kind of revolutionary at the time, because it showed that you could actually build a car that fit together perfectly, and instead of building for volume and building, and then fixing the old mindset, you could actually have a mindset focused on ship it right, on no rework, on customer happiness.

And there was not — this was not a single change. What happened was that people really started thinking about the factories differently. They started engineering a building factories that were engineered to deliver quality, and there's a number of different pieces that I think still resonate today. First is you prioritize quality instead of volume. The second is that you fix the system, not just individual symptoms that you see. The third is that you design for something that can be built, as opposed to designing the wheels and designing the washers, bottles and whatever it is, and hoping it fits together. You solve problems as they occur, as opposed to kicking the can down the road and fixing it at the end. And you empower everybody. Anyone could stop the line, not just the boss.

Now, why am I talking about this? It's actually because modern software development actually embraces all of these principles. If you think about how we build software today, what is considered best practice? A lot of the terms and the practices came from that idea of a Toyota production system, agile, kanban, error budgets if you're in the SRE, well, continuous integration — these ideas all kind of sprung from this notion of how do we build factories that work. And it's worked really well for us in software development so far.

00:02:53 — Code Abundance and the Shift to Agentic Factories

But agents, as we all know, we wouldn't be sitting here if we didn't understand that agents are changing the game. And as agents are changing the game, weird things start to happen. First of all, this is research that was published last month. What of the effect of implementing asynchronous agents in your development lifecycle? And I'm sure many of you have seen this. You have an explosion of the number of lines of code written of 2,450 percent. You have an increase in the number of commits, so the volume co-creation is now easy, and it is now abundant. And by the way, more and more people — I'm sure many people in this room — are saying, okay, if it's easy and abundant, how do I automate it? How do I build factories, agentic software factories, to help automate and scale this? And today we just finished our what we call state of code developer survey. It will be out next month. This is a little preview.

Today, five percent of developers that we spoke to say that they're running agentic software factories, but thirty-four percent said in a year, they expect to win 39 percent in financial services. So people are thinking about how do you actually build these next generation factories, and there's a couple of gotchas in here: security risk. The agents can work really well, help you build stuff. It can also work really well to help you hacks, to help you attacks, to find vulnerabilities, find quality issues, etc. And so, as I was writing this presentation kind of over the weekend — we'll adjust the time again, the Toyota production system — Sachin Adela posted this article next. If you haven't read it, I think it's a great article that I think everyone should internalize. But one of this, he was talking about insider risk of models inside the companies. And one of the things that he talked about was the need for an engineering approach to containment and governance built around strong deterministic system design, human controls, and reliable operating procedures.

And this really resonated with me because what we are seeing is a paradigm shift in software factory design or software development, let's call it more broadly, that we need to engineer — we need to build systems that engineer confidence and quality into identity software factories.

00:05:15 — Cognitive Surrender and Large Codebase Bottlenecks

So to start doing that, I think it's important to understand the limitations that we're facing. So, what's the hard thing about this? And there's five that I'm going to call out. The first is that human in the loop doesn't scale, and this may be — and to this room, it may not be all that controversial, but every — apparently, every company I talk to, I say, "What's your AI governance strategy?" And they give me all the stuff, and they say, but they're worried we have humans in the loop. And the prompt of humans in the loop, as I mentioned earlier, two thousand four hundred and fifty percent increase in lines of code, two hundred and forty percent increase in commits, but the same study showed only a 30 percent increase in releases. We haven't figured out how to scale from producing stuff to shipping stuff.

There's automation needed. There's new tools needed, and really importantly, this is a concept that really I see it every day now, and I'm sure many of you do as well, called cognitive surrender, and this study was done just a couple months ago, and it looked at how often do humans agree with AI, and it turns out they agree with AI when AI is right, 93 percent of the time, and four out of five times, they agree with it when the AI is wrong. So it's not the human in the loop is yeah, we just need to make things faster. It's actually that it doesn't work in the agentic era.

So that's problem one. Problem two is that agents struggle with large, complex codebases. They are very smart, but you need to understand what happens in a code base, in order to work on it. The left-hand side is a study that was done on code translation, and what it shows is that the agents were able to get it right ninety one percent of the time when the codebase was ten thousand lines of code. And by the time it got to even fifty thousand — I'm sure everyone in this room operates on a much bigger code base — when I was trying to translate fifty thousand, the accuracy collapsed.

Similarly, another study looked at how many tokens does it on a large complex codebase to complete the same task, and the variance was 30s, not 30 percent, by 30x, and the reason is because the agent spends a huge amount of their tokens trying to understand how to navigate the codebase, the structure of the codebase, things like this. And it's kind of like if I'm trying to go from L.A. To here, and if I pick a random starting point, if I happen to pick white planes, I will get to New York pretty quickly. If I pick Albuquerque, it's going to take me a long time and a lot of cost. Same thing happens with agents.

00:08:01 — Compounding Errors and Unreliable LLM Judges

Errors compound. This is another important concept. This is study out of meter. And meter looks at, and you can find this on their website — they look at how effective our agents at completing tasks that humans can do, and the way they look at this is: can you complete the task that takes X number of hours? So you can see here that Cloud Windows Preview, which is the latest version that they've published, was good at completing tasks that take a human 16 hours. Okay, so that's really good, but there's an important caveat in the fine print here, which is this is that 50% accuracy. When you add it, when you dial up the accuracy to 80%, it the curve flattens potentially because as you do long-running tasks, errors compound. If you have ninety-five percent accuracy on each task, by the time you're at 20 steps, you have 30 steps, or 36 percent accuracy, right? And so that is another issue we have to deal with in software factors.

Probably the most aggressive headline, but in here is this fourth point that LLMs are unreliable reward-seeking judges. You know, so the question that we always get is, well, can't you just ask an LLM? And so people have been studying this as well, and this is all — again, the first left-hand side is from PayPal AI's research lab. They published a paper this month, actually, and found that depending on the model and the situation, between seven and seventeen percent of verdicts change across what they call identical replications. Ask the same questions, get a different answer between seven and seventeen percent of the time.

And if you start challenging — this is a study out of Meta now — you start challenging the LLM, are you sure? Very simple challenge. It'll actually change this minus seven to thirty-four percent again, depending on the model. So you ask the AI judged twice, and you may get it different answer. And then lastly, and I won't dig into this one too much because I think it's probably self-evident to everyone, as the time to exploit near zero, the ability for us to respond is incredibly challenging. And so prevention is something we believe the software factory has to bake in to the design, as the engineer into the factory. So, as we thought through this, we have written down what we call the code confidence, which is basically how do you build, how do you think about a software factory that delivers what we all want, which is outcomes that we can feel confident in. Right, not outcomes we can trust, not blind trust or anything like that, but outcomes that we can feel comfortable with, that we can sit and tell our executives, our customers, our shareholders, our CEO reports that yes, I believe this factory is working well.

00:10:56 — Four Pillars of the Code Confidence Framework

And we think there's four components to this. The first is being crystal clear about the intent. If you don't know what you're trying to build, you can't tell the agent what to build, and you will not get the other one we want. The second is guide rails, and we very specifically talk about guide rails, not guardrails. Guardrails prevent something from going off the road; guardrails help keep you on it, and I think what agents need are the small nudge to guide and govern those agents.

The third party's verification that needs to be independent and thorough, and issues fixed, as opposed to just hands kicked down the road. And then lastly, evidence: Trust but verify. Trust but understand that it was done with proof that is consistent, audible, and repeatable. And we multiply these together because we've seen a number of situations where one of these elements is missing, and if it's zero, the confidence goes to zero. If you do all of this, you don't have evidence, you go to zero. You don't have confidence in the outcome.

So we really think about factory design as requiring these four elements, and I'm gonna — and as we put together practices, we help customers. I'm not going to walk through every box on this page, but as we help our customers and partners build these software factories, we're very deliberate about how do you get the intent bottom left hand side in purple. How do you put the guardrails in place? How do you do verification in a way that's thorough and independent, and how do you produce the evidence? And by doing this, we see amazing results. So these are real results now from our from our customers. The first one is focusing on intent in guardrails, and in tensing guardrails, we have a product called Sonar Vortex, which gives intent, got goes within the agentic loop, starts to do verifications again, nudging the agents.

And what we see when we provide agents in long-running, large, complex databases with long-running tasks, with this information is a 36 percent reduction in token cost because you're putting it in white claims. You actually are putting it close to the answer. You're helping it understand the map. You don't have to burn tokens on things we already know. And 92 percent reduction in issues in the commit, right? So it's cheaper, and we can get a better outcome to go into that verification stage. This is again all because we're nudging the agent to keep on track.

00:13:29 — Deterministic Verification, SonarQube Results, and Audit Evidence

Then you get to verification, and we started life as a verification company. This is how we see the world, is through a verification lens, and we have a very strong point of view that independent verification has to cover the full range of issues, that is both precise and cost-effective. Right, use CPUs when you can use CPUs, need use GPUs when you need you to. So we look at deterministic mixture of methods, and then you add in a generative mixture of methods, and it comes up with what we think is the gold standard of verification.

And what we see is when you apply this, our product is called SonarQube. So we talked to our customers, over a thousand customers recently, and we said, Are you — how likely are you, or how often have you seen these issues from agentic coding? And the statement to read here is: Sonar users are X percent more likely than non-users to have seen a very positive impact, so 81% more likely to have seen to believe that AI coding has had a very positive impact on code security if you have that gold standard verification. Thirty-four percent more likely to have seen an impact on time to market, seventy-six percent on outage frequency.

So again, as I come back to how do we have confidence to ship, what we're seeing is that if you put these pieces together, and you build verifiable education incredibly strongly as a first-class citizen in the factory, you are actually building confidence, and you're seeing results: better security, more confidence in our time to market, improved or reduction in number of avenues, improved reliability.

And of course, as I mentioned, you have to do this with evidence. Make sure you pass the quality gate. Make sure — whatever there's lots of acronyms up here — whichever ones you need to pass, whether that is your IT general controls inside of your internal controls and operations, whether that's Mister if you're building cars, or any of the other ones that are on here, producing the evidence is what lets you sleep at night. It's what gives you confidence that you can move forward.

So that, in essence, is how we think about the challenge of the future: that code is abundant, that confidence is scarce. And really, the goal that we have and that we think that everyone should have, as we think about the future of software development, is how do we deliver confidence, and the factories of the future need to be able to deliver confidence. So with that, I want to say thank you to everybody. We have a big booth. It's a big red booth on the floor, right over that way, and would love to talk to you more about all of this. Thank you very much. Have a great morning.

Where should AI agents take control? And when should humans stay in charge? Here to explore that balance in policy-driven workflows, please welcome Evan Kormos.