Agent

“I, there'll be more agents than humans in the not too distant future.”

“You've taken like the entire software industry and you've compressed it into this tiny bubble that the only name that people have for it is like agents, but you know that it's going to re-expand again and we're all going to find the lines that exist between th…”

“Our engineers, really what I'm looking for is 10 really good decisions per day, which is very tiring, not like optimizing the random parts of the code. And each one has like, you know, five to 18 agents running at any point in time, doing long horizon tasks on…”

“The reality of the world of tomorrow, which is starting today, is that all of cyber defense, all of software infrastructure, software architecture is moving towards this dynamic state. Where the code will be generated on the fly, the code will be dynamic, it w…”

“This notion of sort of the best user interface ever created is the credit card. This may actually be finally up for renegotiation because agents are in fact smarter than pieces of plastic and even pieces of plastic with rewritable chips.”

“Everyone's going to get hacked because of agents, right? Because the cost of hacking has become almost zero and every surface is going to be attacked.”

“It turns out when agents can run these GTM motions, they will consume 10 to a hundred times more usage than humans ever could. They can run GTM around the clock.”

“The agents have infinite time to research every possible detail. And so they're going to look for where the richest sources of information that are online, which are going to be for the established big companies that are out there.”

“It's been actually a much more challenging problem to build agents within our product than I think I thought it was going to be a year ago to really deliver those outcomes.”

“The agents are going to be a commodity at some point. Your judgment about what to Point them to is the moat”

“There's going to be a new class of citizens on the internet. These are going to be these AI visitors and these AI visitors, these agents are going to book jobs, right?”

“Agent is just a model in a loop with sort of tools and memory and a few other things.”

“Today, a lot of the demand does come from basically, like, the places where people have historically used human panels, we can basically now replace with agents. And these synthetic populations.”

“I think there's still a bunch of experiments going on over there, and it'll take time before people get really comfortable. Unleashing agents in the enterprise.”

“But even then agents are doing code review even better than humans did before.”

“I think about an agent as an LLM call that is wrapped in machinery that's used to construct context. It's really a big context construction machine. And also it provides a way of executing actions.”

“I think in three years, companies will essentially be, like, board of directors, where 20% of the work is the strategy and the thinking, And then agents just do 80% of the execution, right?”

“Getting the agent going is pretty easy, but figuring out where it gets stuck is hard.”

“Every time it has a proposal, like, Hey, we think you should market to these customers in this way. You should send them this email. It will go check its work kind of against its coach that will say, Hey, how good will this, how good do you think this will act…”

“Normally in software, you have this power law of like how good your users are, your products. Do you got a couple of really advanced users? Maybe you put them on a customer panel together. You bring them in and it really wants pushing the limits, but then you …”

“Whether you're building agents, they're going to be the super users, or if you're building the software that you expect, either humans, we're going to plug into MCP or other people are going to expose to algorithms and agents, expect them to be super users. Al…”

“We saw everybody's like, look, by the end of June, you need to be an L three, which is if you're not constantly running either end sessions or a team of agents, right? You're just not, you're not going to be skilled to survive in this next era.”

“There's so few people that are gonna get there in the next year or two. If those skills are gonna be so valuable, if you truly are not talking about AI, right? Yes. But if you're the kind of person that's, I can put an agent to work and actually have to produc…”

“I know this feels, look, a lot of humans, we're used to these kind of fuzzy interfaces, and then we'll meet, we'll have meetings to kind of figure out what he's first going to do. I think that it's not a great way to work, honestly, and it doesn't work with ag…”

“And to me, Twilio is another example. It almost died. It's growing four percent. Now it's growing over 20% because the API worked with agents.”

“We almost like, yeah, we stopped selling. We incentivize all of our sellers and marketing team to just focus on newer products. We still get demand for our older product, but we've also transformed that to agents. So now we are selling agents.”

“The bigger idea is that agents are going to use the internet in a way software uses infrastructure. So they'll request things, they'll call tools, they'll compare products, they'll retrieve data.”

“We spend about the same amount of time, engineer time, tokens, and money on building the evals than building the agents.”

“Coordination of benefits, then you, some of those rules cannot be codified deterministically. You have to make high-level decisions, and that's where the advent of agents has been very helpful, is to manage this deterministic rule sets, in addition to also dea…”

“Inference time compute to perform search over plausible ways of solving the problem that can get much, much higher performance or much more reliability in, Long-running agent flows.”

“Particularly in the internal Google development environment, we have skills so that the agents can know how to use lots of our internal tooling for coding, or for code reviews, or for, you know, measuring performance, or, you know, fetching log files, and thos…”

“You will be using a software factory. Like, and what I mean is, the agents are going to be writing a hundred percent of your code, reviewing a hundred percent of your code, shipping a hundred percent of your code. Like, that's where we're going.”

“So like the old way of automating workflow type tools, it was very Specified declarative in the way that those steps were laid out, and if anything deviated from that, it had to have exception processing. Now with agents, you don't need to be as definitive. Th…”

“If you have an agent that's tasked with doing a job, If its goal for existing is to complete that job, but it doesn't have what it needs to complete that job. It's not going to error out. It's not going to time out. It's going to probably hallucinate what it n…”

“It's going to be a, it's going to be an org chart of agents passing work off to it.”

“I think some of the biggest insurers will struggle. They will not be able to keep up from a margin standpoint with where we can get to with agents. I think some of them do have Technical expertise. It's more operational and kind of people ops. If you've built …”

“I still think that we will be a few months hence before folks are actually agreeing to You know, agents using the wallet beyond a particular scoped transaction”

“I do think engineering organizations are going to evolve. I don't know exactly what it looks like yet, but I think you will increasingly have probably somewhat smaller teams of higher caliber and higher quality engineers, because I think a lot of the lower lev…”

“There's a platform sense in which I think there will be, will have, people will have an agent, maybe a couple, but certainly an agent that's kind of their general front door.”

“I think agents are the new SaaS because software is moving from help me do the work to do the work with me.”

“From my perspective, the way to start in a business introducing agents to a company is to give each person their own agent.”

“Initially, one of the biggest mistakes that people make is that they just want to throw something in an agent, and then walk away and expect to get a finished result at the end.”

“It's not that management is going away. It's not that everyone's an IC, but like everyone's kind of both now, right? If you're an IC, you're not typing code out character by character, right? Like you are managing something. You're managing agents. You're mana…”

“Agents are really good at ads because they have, like, endless ideas, whereas I feel like as humans, if you want your ROAS to be really high, you need to change it, like, every other day or so and change it, like, seven times over.”

“If you start to talk to the agent in this builder in the back end, it will develop its own, what I call personality or you can call it brain, but it will develop its own personality to do it.”

“The nice thing about building something like an agent is any ideas it's going to give you is only based in data. Like, that's all it has. It's got all your data. It's got your numbers. It's got a goal, hopefully. You give it a goal, maybe write down your goal …”

“Now with agents, they're becoming very CLI first. And so this one's kind of obvious to everyone, but you know, what spills out the other side of this is, okay, what that means is that everything is very code first. It's very Git first.”

“If agents are building now, then they'll probably be operating in the future. So, at the moment, maybe developers end looking at things, whether the database is healthy or not. But of course, they probably don't want to be doing that, and that's actually where…”

“Plus the agent is like completely useless if you can't share sessions with someone and have history and have search and all this like layer on top of it for collaboration.”

“Instead of worrying about PostgreSQL and maybe Spark SQL, why not just one? But I don't think that's an issue for agents. Agents are very eloquent in PostgreSQL or Spark SQL. It's never going to get confused. As long as the data is there and it's accessible ag…”

“Actually, I think many of the traditional software will be sort of rewritten with this new paradigm, which is just get the data to be there. And then they slap some agent on top.”

“Now, things that are that simple, to be clear, are covered at this point by most agents, right? You know, they all They, despite some issues, yeah, normal, normal sort of, you know, will not be that easily fooled by just push all my API keys to a public thing,…”

“If you parse some untrusted content and there is like a prompt injection, you know, something that's clearly trying to get the model to do a bad thing, you might be interested in knowing about that, but you don't necessarily like want your cloud code that you …”

“Six to 12 months ago, the approach was just jam as much information into whatever system you can. And we were kind of promised that it would just work and AI is just gonna, and like, there's, you know, you'd get this feeling that, oh wait, a hundred page PDF. …”

“Agent loops, agent harnesses are all running on CPUs. The tokens are produced by the frontier models on GPUs, but whatever work, like, let's say like Claude generates a coding script that decides to download 500 files from different websites and then You know,…”

“The advertising model around like travel or shopping or like fashion are not getting disrupted by agents because the judgment is not objective. Any, anything where the judgment is objective, the transaction is based on objective judgment, that's going to get d…”

“Like, normal humans? No. I mean, I'll use them.”

“Well, I can report from the singularity as someone who's building agents every day that these things are way far away.”

“And we've grown to accept that, that, you know, we're all here in the Valley with a plan. To automate white collar work such that there's going to be mass unemployment, according to these folks who are totally wrong, in my opinion, in three years, and all the …”

“Coding has worked so well that we now have to rebuild the IDE, right? I mean, it's kind of nuts to see what we launched is like, oh my God, I have these hundred agent sessions. I, the cognitive load, it transfers back to me as a human is so excessive that now …”