Building the Real-World Infrastructure for AI, with Google, Cisco & a16z

Summary

a16z Podcast: Building the Real-World Infrastructure for AI, with Google, Cisco & a16z

Episode Overview

This episode of the a16z Podcast, recorded on October 29, 2025, dives into the massive shift underway as artificial intelligence moves from software to transforming global physical infrastructure. Host Andreessen Horowitz is joined by Amin Vadat (Google), Jeetu Patel (Cisco), and Raghu Raghuram (a16z) to unpack how the unprecedented scale and urgency of AI are pushing limits on chips, power, data centers, and networking. The conversation covers the challenges of building next-generation infrastructure, the ripple effects on geopolitics and economics, and what startups and enterprises should anticipate for the coming year.

---

1. The Infrastructure Renaissance: Why AI Has Changed the Stakes

Key Points:

• AI Infrastructure vs. The Internet Boom: Panelists agree the current AI-driven infrastructure build-out dwarfs previous technology cycles—even the Internet’s expansion in the late 90s/early 2000s.
  • Quote:
    "This is like the combination of the build out of the Internet, the space race and the Manhattan Project all put into one." – Jeetu Patel (00:00, 02:32)
• Unprecedented Scale and Urgency: Demand for compute, power, and data center infrastructure is at historic highs; projections are likely underestimating future needs.

Quote Highlights:

• "10x is an understatement. It's 100x what the Internet was." – Amin Vadat (02:08)

Timestamps:
00:00–03:49: Framing the infrastructure surge and its historic parallels
02:07–03:21: Expert perspectives on prior cycles versus today

---

2. Demand, Constraints, and Global Implications

Planning for Scale

• AI Demand Is Unmatched: Google’s custom AI chips (TPUs) have 100% utilization, demand routinely outpaces supply, and legacy hardware is stretched to its limits (03:49–05:22).
  • "Our seven and eight year old TPUs have 100% utilization. That just shows what the demand is." – Amin Vadat (03:49)
• Bottlenecks: Core constraints are around power, land, permitting, and supply chain—not just chips.

Enterprise vs. Hyperscale Readiness

• Enterprises Lag Behind: Most enterprise data centers need total redesigns for power and rack requirements, suggesting years of lag compared to hyperscalers and “NeoClouds” (05:42).
• Geography of Power: Data centers are now being built where power is available, not where companies want them for convenience, creating new global patterns (06:42–07:43).

---

3. System and Networking Paradigms for the AI Era

From Commodity PCs to Co-Designed Stacks

• Not a Return to Mainframes: While Nvidia’s "mainframe-like" clusters are influential, the dominant pattern remains software-scale-out on giant pools of specialized hardware.
  • "We're going to be reinventing computing…and five years from now, whatever the computing stack is…is going to be unrecognizable." – Amin Vadat (08:14)
• Co-Design Is Key: Software and hardware evolve together, just as Google’s early systems did; integrated system design will define this era.

Industry Collaboration and Open Ecosystems

• Tight Integration Across Layers: Success now rests on deep, sometimes months-long design partnerships between companies (10:07).
  • "We will have to work like one company, even though we might actually be multiple companies." – Jeetu Patel (10:07)

---

4. Hardware and Specialized Processors: The Golden Age

Boom in Specialization

• Beyond GPUs: Future lies in highly specialized processors for particular workloads (GPUs, TPUs, other custom silicon). Time to market for specialized hardware is a current limiter (11:40–12:51).
  • "We’re really seeing the golden age of specialization…For certain computation [TPUs are] somewhere between 10 and 100 times more efficient per watt." – Amin Vadat (11:40)
• Geopolitics of Chips: Divergence between China and the US—China emphasizes scale with less advanced manufacturing; the US focuses on cutting-edge, high-efficiency chips (13:24).
  • "This will actually have a really interesting implication on geopolitical structures as well." – Jeetu Patel (13:24)

---

5. The Reinvention of Networking

Networking as a Force Multiplier

• Bottleneck Alert: The network is emerging as a leading performance constraint in data centers; bandwidth demands are soaring (15:25).
  • "The network is becoming a primary bottleneck, which is scary." – Amin Vadat (15:25)
• Novel Architectures: Transforming for both “scale up” (within a data center) and “scale out/across” (connecting far-apart centers as logical units). Networking must accommodate extremely bursty, unpredictable workloads (17:00).
• Silicon Diversity: Cisco aims to disrupt single-vendor silicon dominance, promote alternatives to Broadcom for greater flexibility and efficiency (19:48).

---

6. Inference, Training, and Future Use Cases

Specialized Architectures

• From Training to Inference: New, more specialized infrastructure is being deployed for inference vs. training. Reinforcement learning and latency-optimization are growing priorities (20:12).
  • "We are deploying specialized architectures for inference and I think as much software as hardware." – Amin Vadat (20:12)
• Reducing Inference Costs: Great leaps in efficiency (10x–100x) are being achieved, but user demand for better quality models keeps raising computing needs (20:59).
  • "It's almost a never ending loop where you'll need to have more performance for inference in perpetuity." – Jeetu Patel (21:54)

---

7. Real-World AI Usage: Inside Google and Cisco

Internal AI Applications

• Code Migration at Scale: Google recently used AI to migrate massive, multi-language codebases from x86 to ARM, achieving what previously would've taken “seven staff Millennia.” (23:20–24:41)
  • "We decided to tell the company, hey, everyone needs to move from bigtable to Spanner... the estimate for doing that migration for Google was seven staff Millennium." – Amin Vadat (24:01)
• AI in Practice at Cisco:
  • Coding use cases (migration, debugging, greenfield projects)
  • Sales prep, legal contract review, product marketing drafts
  • Cultural shifts: Engineers must re-evaluate tools continuously as improvement is so rapid (25:18–28:02)
  • "You have to come back to it within four weeks and see if it works again. Because the speed at which these tools are kind of advancing is so fast..." – Jeetu Patel (27:08)

---

8. Looking Ahead: Advice for Startups & Roadmaps

For Founders:

• Go Beyond Wrappers: Building simple wrappers around AI models is a fragile business—true value comes from close integration between models, products, and user feedback loops (29:17).
  • "Don’t build thin wrappers around models that are other people’s models. …the durability of your business will be very, very short lived." – Jeetu Patel (29:17)
• Prepare for AI-Driven Productivity: Multi-modal models (video, images, etc.) are on the cusp of the breakthroughs seen in NLP, with huge productivity and educational opportunities (30:48).

---

Memorable Moments & Notable Quotes

• "Seven staff Millennia." – Amin Vadat, on the scale of Google’s code migration challenge (24:30)
• "We will have to work like one company, even though we might actually be multiple companies that actually do these pieces." – Jeetu Patel (10:07)
• "We're grossly underestimating the build out. I think there's going to be much more needed than what we are putting the projections towards." – Jeetu Patel (02:32)
• "If you have low latency and low performance and high energy inefficiency, then every kilowatt of power you save moving the packet is a kilowatt of power you can give to the GPU." – Jeetu Patel (17:53)

---

Segment Timestamps for Key Topics

• 00:00–03:49 — State of AI infrastructure and historic parallels
• 03:49–05:22 — Current cycles of capex, supply, and demand
• 05:42–07:43 — Differences between enterprise and hyperscaler infrastructures
• 08:14–11:09 — The future of systems, hardware-software co-design
• 11:40–14:47 — Next-generation silicon, geopolitical considerations
• 15:25–19:48 — Networking as a bottleneck, diversity in components
• 20:12–22:53 — Training vs. inference, cost reductions
• 23:17–28:02 — How AI is used internally at Google and Cisco
• 28:30–31:28 — Advice for startups, upcoming industry shifts

---

Summary

The episode delivers a grounded, provocative perspective on how AI is driving the largest infrastructure build-out in decades. The panelists emphasize that solutions for compute, power, and networking must be deeply reimagined, not simply scaled from previous models. As companies like Google and Cisco rapidly adapt internal processes and reorganize priorities, the panel advises founders to look for opportunities in tight integration—both across teams and architectures—and warns that the next year’s advances may soon make even today’s innovations seem routine.

Transcript

[A]

The good news is infrastructure is sexy again. So that's kind of cool. This is like the combination of the build out of the Internet, the space race and the Manhattan Project all put into one where there's a geopolitical implication of it, there's an economic implication, there's a national security implication, and then there's just a speed implication. That's pretty profound.

[B]

I mean, I think it's easy to say I've seen nothing like this. I'm fairly certain no one's seen anything like this. The Internet in the late 90s, early 2000s was big and we felt like, oh my gosh, can't believe the build out, the rate this makes it. I mean, 10x is an understatement. It's 100x what the Internet was.

[C]

The AI boom isn't just changing software, it's transforming the physical infrastructure that runs it. Today you'll hear a conversation with Amin Vadat from Google, Jeetu Patel from Cisco, and Raghu Raghuram from A16Z on what it takes to build the real world systems behind large scale AI, from chips in power to data centers and networking. They discuss the scale of the current build out, the new constraints on compute power in interconnect, and how specialization in hardware and architecture is reshaping both the industry and global geopolitics. It's a grounded look at how infrastructure itself is being reinvented for the AI era and what comes next. Let's get into it.

[D]

What better time and place to talk infrastructure? All right, so we were back in the green room and just as the first question was getting answered, I got cut off. So this could be an entire repeat for all in all. But anyway, let's go. Right, the first question is similar. Both of you firstly welcome and thank you for being here. Thank you. Hope you'll have a great day and a half as well. Both of you been in the industry for a while and both of you have lived through many infrastructure cycles, right? So have you seen anything like this cycle from your vantage point? Not from an investor vantage point, but from your internal vantage point where you are responsible for building things and planning for things and so on. Any one of you, where do you want to start? You want to start? Amin?

[A]

Go ahead, Amin.

[B]

I mean, I think it's easy to say I've seen nothing like this. I'm fairly certain no one's seen anything like this. The Internet in the late 90s, early 2000s was big and we felt like, oh my gosh, can't believe the build out, the rate this makes it, I mean 10x is an understatement. It's 100x what the Internet was. I think the upside is as big as the Internet was. Same thing, 10x and 100x. Yeah, nothing like it.

[A]

Yeah, I agree. I don't think there's any priors to this size, the speed and scale. I'd say the good news is infrastructure is sexy again so that's kind of cool. It was a long time where it wasn't sexy. The thing I would say that's really interesting is this is like the combination of the build out of the Internet, the space race and the Manhattan Project all put into one where there's a geopolitical implication of it, there's an economic implication, there's a national security implication and then there's just a speed implication that's pretty profound. So yeah, none of us have ever seen it at this size and scale. On the other hand, I think we are grossly underestimating. The most common question I asked right now is is there a bubble? I think we're grossly underestimating the build out. I think there's going to be much more needed than what we are putting the projections towards.

[D]

So that's the follow on question. Where are we do you think in the capex but more importantly what are the signals that you guys use internally right in your thinking? I mean you have to plan data centers, whatever four or five years in advance, you have to buy nuclear reactors and whatnot. So how do you think about the demand signals as well as your technology signals And Jeetu, the same thing for you but from the point of view of Enterprise and NeoCloud et cetera.

[B]

Amit, we're early in the cycle is what I would say certainly relative to the demand that we're seeing. Our internal users are We've been building TPUs for 10 years so we have now seven generations in production for internal and external use. Our seven and eight year old TPUs have 100% utilization. That just shows what the demand is. Everyone of course prefer to be on the latest generation but whatever they can get. So this tells me that the demand is tremendous but also who we're turning away and the use cases that we're turning away it's not like oh yeah, that's kind of cool, it's oh my gosh, we're actually not going to invest in this and there's no option because that's where we are on the list. Same with many of you in the room we're working with many of you in the room and many of yours are telling me directly and thank you. We need more earlier now. The challenge here though is, as you said, we're limited by power. We're limited by transforming land, we're limited by permitting, and we're limited by backup delivery of lots of things in the supply chain. So one worry I have is that the supply isn't actually going to catch up to the demand as quickly as we'd all like. I heard in the previous session some of the discussions of the trillions of dollars that we're going to be spending, which I think is accurate. I'm not sure that we're going to be able to cash all those checks. In other words, literally you all have some money, you can't spend it all as fast as you want. I think that's going to extend for three, four, five years.

[D]

Wow. And how do you deal with the depreciation cycles that are involved there? Does the demand curve and the depreciation cycle curves match up?

[B]

Well, fortunately we buy just in time. But the nice thing is just in time for the hardware. The depreciation cycle for the space power is more like somewhere between 25 and 40 years. So we have benefits there.

[A]

I think if you think of on the networking side and you look at both enterprise and the hyperscalers as well as NEO clouds, I think the story is quite different. So the enterprise is pretty nascent and it's built out of true infrastructure. I just don't think that the data centers, like if you assume that 100% of the data centers at some point in time will need to get re racked and you will need a very different level of power requirement per rack that's going to be there compared to what used to be there in the traditional data centers. I just don't think that the enterprises are far enough along. Maybe the few enterprises that are at super high scale might be there, but I don't think the enterprises are far enough along. Hyperscalers and NEO clouds is a completely different story. And to Amin's point on this notion of scarcity of power, compute and network being the three big kind of constraints in this thing, I would say right now that because there's not enough power singularly in one location, data centers are being built where the power is available, rather than power being brought to where the data centers are. And that's why you're seeing a lot of projects that are being built out all throughout the world. The other point though is the lion's share of the constraints that we're going to have I think are going to be sustainable for a long period of time. And as you have data centers that are being built farther and farther apart, one there's going to be a huge demand for scale up networking so that you can have a rack that gets more and more networking for scale up. The second is you're going to have a lot of demand for scale out where you have multiple racks and clusters that need to get connected together. But we just launched a new piece of silicon as well as a new chip and a system for scale across networking where you might have two data centers that act as a logical data center that could be a up to 8, 900 kilometers apart. And you will see that just because there's not going to be enough concentration of power in a single location. So you'll just have to have different architectures that get built out.

[D]

Actually that brings us to the next topic that I want to discuss. The future of systems and networking and so on and so forth. So Google bought the first or at least large scale scale out commodity servers and production for the web revolution and now Nvidia is bringing back the mainframe in a different form. So what do you think happens next? I mean is this the new style of coherent cluster wide computing that we need and there's going to be shared memory and all sorts of things or do you think the pattern changes?

[B]

Again, I don't think we're quite too back to mainframes in that it is still the case that people are running on scale out architectures across these pools. In other words, Whether you have GPUs or TPUs, you're not necessarily saying hey that's my GPU supercomputer. You're saying I've got 16,384 GPUs and maybe I'm going to go grab some subset. Now I've got uniform all to all connectivity in many cases, which is fantastic. Same with TPU's. It's not like I say I have a 9000 chip pod and I have to make my job fit on that. Maybe I actually only need 256, maybe I need 100,000. So I do think that actually this software scale out is still going to be there. I'll note two things though. One, you're absolutely right that say about 25 years ago at Google and other places simultaneously there was really a transformation of computing infrastructure. Like the notion that actually you would scale out on commodity PCs, essentially the same ones that you could Buy off the shelf running a Linux stack. And that's what you would do for disk, that's what you would do for compute, that's what you do for networking. I mean you all take it for granted that this is sort of, it was radical. There are many people who thought that this was a terrible idea that wasn't going to work. I think the exciting thing about this moment right now is actually that we're going to be reinventing. I'm not saying Google, we are going to be reinventing computing and five years from now, whatever the computing stack is from the hardware to the software is going to be unrecognizable. And by the way, there was this co design because if you think about it, I'll use Google examples because I know those best bigtable, Spanner, gfs, Borg, Colossus, they were hand in hand co designed with the hardware, the cluster scale out architecture. And we wouldn't have done the scale out hardware if you didn't have the scale out software. Same thing is going to happen in this moment. So I think actually the mainframe is going to look very, very different.

[A]

I do think there'll be this extreme demand for an integrated system because right now we are very fortunate at Cisco where we do everything from the physics to the semantics. You think about the silicon to the application. And other than power, one of the constraints is how well integrated are these systems and do they actually work with the least amount of lossiness across the entire stack. And so that level of tight integration is going to be super important. And what that means the industry will have to evolve into is we will have to work like one company, even though we might actually be multiple companies that actually do these pieces. And so when we work with hyperscalers like Google or others, there's a deep design partnership that actually goes on for months and months together ahead of time before we actually even do the deal. And then once the deal is done, of course there's a tremendous amount of pressure to make sure that they're moving pretty fast. But I think the industry's muscle of making sure that you operate in an open ecosystem and not be a walled garden is going to get important at every layer of the stack really.

[D]

Agree. Let's talk about the disaggregate the stack a little bit. One of the most interesting topic is processors, right? Clearly there's an amazing vendor producing an amazing processor that has massive market share today. And we see startups all the time doing all sorts of processor architectures. You got an amazing processor inside Your fortress. What do you think happens next in processor land?

[B]

Yeah, we're huge fans of Nvidia. We sell a lot of Nvidia products and chips. Customers love them. We're also huge fans of our TPUs. I think the future is actually really exciting and actually I don't think that we've hit the point of okay, there's TPUs, there's GPUs, there's whatever trainiums or something else. We're really seeing the golden age of specialization. And that's my observation. In other words, if you look at it a tpu. I'll use that example again because I know it best. For certain computation is somewhere between 10 and 100 times more efficient per watt. And it's this watt that really matters than a CPU that's hard to walk away from 10 to 100x. And yet we know that there are other computations that if you built even more specialized systems for but not just a niche computation computations that we run a lot of at Google, for example, maybe for serving, maybe for agentic workloads that would benefit from an even more specialized architecture. So I think that actually one bottleneck is how hard is it and how long does it take to turn around a specialized architecture. Right now it's forever.

[D]

Yeah.

[B]

For the best teams in the world, really. From concept to live in production, speed of light is two and a half years.

[D]

Yep.

[B]

I mean that's, that's if you nail everything right. And there are a few teams that do. But how do you predict the future? Two and a half years out for building specialized hardware? So A, I think we have to shrink that cycle. But then B, at some point when things slow down a little bit, and they will, I think we're going to have to build more specialized architectures because the power savings, the cost savings, the space savings are just too dramatic to ignore.

[A]

And this will actually have a really interesting implication on geopolitical structures as well. Because if you think about what's happening in China, China actually doesn't make 2 nanometer chips. They make, you know, 7 nanometer chips. And so if you think about what but they have unlimited amount of power and they have unlimited amount of engineering resource. And so what they can do is do the optimization on the engineering side, keep the 7 nanometer chips and make sure that they give people unlimited amount of power. We might have a different architectural design where you have to get extremely power efficient. You don't have as many engineers as you might enjoy in China and you can actually go to 2 nanometer chips and those might be power efficient in some ways, but they might have thermal lossiness in other ways. There's a whole bunch of things that have to get factored in on the architecture that will get more specialized even by GEO and by region and then depending on how the regulatory frameworks evolve, how that GEO then expands. If China expands to different regions in the world, you will have a very different architecture that plays out than if America expands to different regions in the world. So this is a very interesting kind of game theory exercise to go through on what happens in the next three years in tech in general. And no one knows right now.

[D]

That's the beauty of the world that we live in.

[A]

Yeah, yeah.

[D]

So we'll soon be measuring systems by engineers per token in addition to watts per token. All right, so let's turn to another.

[A]

Topic which varies engineer per kilowatt, engineer.

[D]

Per kilowatt in the U.S. networking. Right. Obviously you alluded to it scale up, scale out. In your case, you mentioned scale across. So it seems to me that networking is also going to get reinvented in a fairly significant way. So what are the leading signs that you're seeing and the signals that you're seeing on the direction networking is going to take?

[B]

Yeah, networking is going to need a transformation for certain. In other words, the amount of bandwidth that's needed at scale within a building is just astounding. I mean, and it's going up. The network is becoming a primary bottleneck, which is scary. So more bandwidth translates directly to more performance. And then given that the network winds up actually being a small power consumer that delivered utility, you get per watt, like it's a super linear benefit, like spend a little bit here, get way more there. So I think that that side is absolutely there. I'll put in a plug here in that for these workloads we actually know what the network communication patterns are a priori. So I think this is a massive opportunity. In other words, do you then need the full power of a packet switch when actually you know what the rough circuits are going to be? And I'm not saying you need to build a circuit switch, but there is an optimization opportunity. The other aspect of this here is these workloads are just incredibly bursty and to the point where, and we've written about this, power utilities notice when we're doing network communication relative to computation at the scale of tens and hundreds of megawatts, like massive demand for power, stop all of a sudden and do some network communication and then burst back to computing. So how do you build a network that needs to go at 100% for a really short amount of time and then go idle?

[D]

Yeah.

[B]

And then same actually for the scale across use case, which we're absolutely seeing. You don't run large scale free training across all your wide area Data center sites 12 months of the year. So. And then you're going to. This is a problem I think about a lot is let's say you build the latest, greatest chips in these three data center sites. How long are you going to be there before you migrate to the latest, latest chips and 3 other sites? And then what do you do with the network that you left behind? People are going to run jobs on them, but you're not going to need nearly the network capacity that you did for large scale training retraining anyway. So the shift of needing massive Networks for like 5% of the time. I don't know how to build a network like that. So if any of you do, please, please, please let me know.

[D]

I mean, if you don't know how to build this, there's nobody that knows.

[B]

How to build this. We're trying to figure it out. It actually is a fascinating problem.

[D]

Yeah.

[A]

I do think like if you think of if power is the constraint and if compute is the asset, I think network is going to be the force multiplier. Because if a packet, if you have low latency and low performance and high energy inefficiency, then every kilowatt of power you save moving the packet is a kilowatt of power you can give to the gpu, which is super important. The other thing is when you think about scale up versus scale out versus scale across, you'll also need, especially on inference versus training, there are different things that get optimized. You might optimize for latency much more on training runs. You might optimize much more for memory on inferencing. There's architectural. And so I also feel like the way that networking will evolve is rather than it being a training infrastructure that then gets applied to inferencing, you might have inferencing native infrastructure that gets built over time. There's good considerations to look at on how all of the architectural components are moving. But in my mind, if I were to say strategically, one of the biggest things that's happening in networking from our vantage point is if you're just a wrapper around Broadcom, then you've got a monopoly that's going to be a very predatory one. And so one of the big reasons where Cisco is super relevant Is you don't just have a Broadcom world with people just wrapping Broadcom that their systems are on Broadcom, but you will actually have a choice of silicon and that choice and diversity of silicon is going to be super important, especially for high volume kind of consumption patterns.

[D]

So last question on the system since you brought that up and we'll move to use cases. Inference both of you mentioned, you talked about in the context of the processors. You just started talking about the architecture. Are you deploying today's specific architectures for inference? I mean, is it still shared workloads?

[B]

We are deploying specialized architectures for inference and I think as much software as hardware, but the hardware is also deployed in different configurations is the way I would say it. And then the other aspect of inference that is becoming really interesting is reinforcement learning, especially on the critical path of serving because latency just becomes absolutely critical. And I think that so how you would build your system and how you would connect it up to one another and of course networking plays a key role there becomes increasingly interesting.

[D]

And are there singular choke points that if removed would accelerate the thousand fold reduction in the cost of inference that we need or is this just a natural curve that we are writing down?

[B]

So we're massive. I mean two things here, One again maybe many of you are familiar with this. Pre fill and decode on inference look very, very different. So actually ideally you would have different hardware. Actually the balance points are different. So that's one opportunity. It comes with downsides. We can talk about that. What I would say though is that maybe something people don't realize is that we're actually driving massive reductions in the cost of inference. I mean 10 Xs and 100 X's. The problem or opportunity is the community. The user base keeps demanding higher quality, not better efficiency. So just as soon as we deliver all the efficiency improvements we're looking for, the next generation model comes out and it is the whatever intelligence per dollar is way better, but you still pay more and it costs more relative to the previous generation. And then we repeat the cycle and.

[A]

It'S almost like the longer the reasoning that you have, the more impatient the market gets. For example, if you have a 20 minute reasoning cycle, like for example with deep research, you could have autonomous execution for about 20 minutes. That was interesting. Now you have most of the coding tools that can go up to 7 hours to 30 hours of duration of autonomous execution. When that happens, there's actually a greater demand for saying compress that time down. And so it's kind of a Self fulfilling prophecy where you need to have more performance because of the fact that you've been able to go out and do things for a longer autonomous amount of time. And so it's almost a never ending loop where you'll need to have more performance for inference in perpetuity.

[D]

Yeah. Though intelligence per dollar is a business model metric, so it is not just a processor capability.

[B]

No, it's end to end. Absolutely.

[D]

Yeah. Okay, so let's change topics and talk about actual usage. So both of you have massive organizations. Where are the key wins that you're getting today with applying all the AI that's available to you. Then we'll talk about what your customers are doing. But I'm actually curious about what you're.

[B]

Doing internally within the teams.

[D]

Yeah.

[B]

So I mean coding is the obvious one and that's actually picking up increasing traction and increasing capability. We just actually in the last couple of days published a paper that showed how we applied AI techniques to do instruction set migration. So in other words, we actually had a fairly massive migration from x86 to ARM, making our entire code base. And at Google it's a very, very large code base, sort of instruction set agnostic and including to, you know, future RISC V or whatever else might come along, tens and thousands, hundreds of thousands of individual entire code base.

[D]

You're going to make it agnostic entire.

[B]

Code base because we want and need all of our code base to be.

[D]

That's a crazy ass project.

[B]

Yeah, so it was. And the motivation though for this actually was a few years ago we had this amazing legacy system called BigTable and then a new amazing system called Spanner. And we decided to tell the company, hey, everyone needs to move from bigtable to Spanner. And by the way, bigtable was amazing for its time, but Spanner was better. The estimate for doing that migration for Google was seven staff Millennium.

[D]

How much? How much?

[B]

Seven staff Millennia. We had a new unit that we had to actually to see what. And it wasn't like made up people being lazy. It's like this is what it was.

[A]

It's endearing that they came up with that though.

[B]

And you know what we decided, Long live bigtable. I decided what? It just wasn't worth it, honestly. The opportunity cost was too high and we have these sorts of migrations, TensorFlow to Jax. We actually, I mean again, somewhat private but not too secret. We've affected this internally with AI assist, went integer factors faster. Now there are other tasks which the tools probably aren't quite yet up to the whatever standard for but the area under the curve is getting bigger and bigger and bigger.

[A]

So we are seeing probably like three or four really good use cases and then we are seeing some use cases which are not working yet. And so what is working? Code migrations are working relatively well so far. We use largely a combination of Codex, Claude and Cursor, some Winsurf. And so code migrations tends to work pretty well. Debugging, oddly enough, has actually been very, very productive with these tools, especially with clis where we've not done as good a job. And Then front end 0 to 1 projects tend to do extremely well. Like the engineers are super productive when you go to code that's older and especially further down in the infrastructure stack, much harder to go out and get that to happen. But the challenge that we have to orient our engineers on this is actually much more of a cultural reset problem than it is just a technical problem, which is if someone uses something and says this isn't working right, you can't put it back on the shelf saying this doesn't work for another six or nine months. You have to come back to it within four weeks and see if it works again. Because the speed at which these tools are kind of advancing is so fast that you almost have to kind of get like I was with 150 of our distinguished engineers today and what I had to urge them to do is assume that these tools are going to get infinitely better within six months and make sure that you get your mental model to where that tool is going to be in six months. And what are you going to do to be best in class in six months rather than assessing it for where it is today and then putting it aside for six months, assuming that that's not going to work for the next six months. I think that's a big strategic error. So we've got 25,000 engineers. I'm hoping that we can get at least 2 or 3x productivity within a very short amount of time within the next year. And we'll be able to see if that happens. A couple other big areas that we are starting to see some good responses is in sales preparation, going into an account call, really good legal contract reviews, actually much better than what we had thought. And then the last one is not super high inference volume, but product marketing. I think the first ChatGPT take on competitive is always better than what any product marketing person comes up by themselves. So we should never start from right slate. Just start from ChatGPT and then go from there.

[D]

Okay. We could be talking about the topic for a long time but they showed me the two minute warning. So I want to focus on one last question here. So we've got a lot of founders here, right, Building amazing companies. So what is the most interesting development they should look forward to in the next calendar year, let's call it, or the next 12 months a from your company and B from the industry. If you are looking at your crystal ball.

[B]

I think to build on the point, these models are getting more spectacular by the month and then there'll be from whatever companies you like, a bunch of really exciting, including ours.

[D]

I forgot to say you're not allowed to say models will get better.

[B]

Everybody knows the models are going to get but I mean they're getting scary good is the part that I would say. But I think that then the agents that get built on top of them and the frameworks for making that happen are also getting scary good. So the ability to have things go quite right for quite long over the coming 12 months is going to be transformative.

[D]

Do you want to leak any aspect of your roadmap next 12 months?

[B]

Not right now. Yeah.

[D]

Okay. Do you too?

[A]

I'd say the big shift and what I would urge startups to do is don't build thin wrappers around models that are other people's models. I think the combination of a model working very closely with the product and the model getting better as there's feedback in the product is going to be super important. So you are going to need foundation models but if you just have a thin wrapper, I think the durability of your business will be very, very short lived. So that would be something that I would urge you on and I think the intelligent routing layer of some sort that says I'm going to use my models for these things, I'm going to probably use foundation models for other things and dynamically keep optimizing will be, I think cursor does that pretty well. But that'll be a good way that the software development lifecycle will evolve. What you should expect from Cisco is look, truth be told, for the longest time people thought Cisco was a legacy company. There has been, I think there's a level of momentum in the business, there's a spring in the step in the employee base. So you should expect, like I said, from the physics to the semantics in every layer, from silicon to the application, a fair amount of innovation in silicon and networking and security and observability and the data platform as well as applications from us. And we're excited to work with the startup ecosystem. So if you ever feel like you want to work with us, make sure that you reach out to us.

[D]

You got to say something, Amit.

[B]

I mean, one aspect that I want to highlight about the models is where we were with, let's say, text models two and a half, three years ago. They were fun, like, hey, write me a haiku about Martin did a great job. Now they're amazing. I think that what's going to happen in the next 12 months is the same thing is going to be happening with input and output of images and video to these models. And to the extent that even for images, imagine them as productivity and educational tools. Not just okay, here's Martine as Superman on a like, that's cool too, right? But using it for productivity gains and learning I think is going to be really, really transformative.

[D]

Awesome. So on that note, we'll end this session. Thanks for a great conversation. Amin thanks G2.

[C]

Thanks for listening to this episode of the A16Z podcast. If you liked this episode, be sure to like, comment, subscribe, leave us a rating or review, and share it with your friends and family. For more episodes, go to YouTube, Apple Podcasts, and Spotify. Follow us on X16Z and subscribe to our [e-mail] thanks again for listening and I'll see you in the next episode. As a reminder, the content here is for informational purposes only. It should not be taken as legal, business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any A16Z fund. Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see a16z.com disclosures.