Aravind Srinivas
Co-founder and CEO of Perplexity, the AI answer engine company that has expanded into autonomous agents, research and knowledge-work tools.
“As this competition at the model layer arises, the value accrues to the one who can orchestrate all of these in a really solid harness that stays model agnostic, lightweight, so you're not building around the limitations of models. You're actually benefiting from the progress in the models.”
Newcomer Pod · Startups & Venture · November 2026
Context
Newcomer asked whether companies building on top of other labs' models can thrive. Srinivas bets on a world of several competing models where the value sits in orchestrating them.
Transcript
…the underlying infrastructure of powering these agents. We built our own sandboxes, we built our own model routing, and we host our own models when we delegate to open weight models that are fine-tuned. And so this is the way in which we get to be the orchestrator. of all possible knowledge work.
…for companies like yours now is to just be so much better at building harnesses that there's inherent value just in that and not have to basically let the foundation model companies build the models, but they just suck at harnesses and product and they're just not going to be threatening to you?
We believe in the world of multiple models. Just as Anthropic's making great progress, so is Google. Carbon is pretty much at fable quality. And OpenAI is behind today, but that's not going to be the case for too many weeks or months. And so as this competition at the model layer arises, the value accrues to the one who can orchestrate all of these in a really solid harness that stays model agnostic, lightweight, so not a lot of hard coding. So you're not building around the limitations of models. You're actually benefiting from the progress in the models. And world-class infrastructure. You need to hold your own sandboxes. You need to keep it safe. Enterprises need to be able to trust their agents to be running on your platform. It needs to respect role-based access. Agents cannot go rogue inside your sandbox. So all that stuff matters tremendously. Connectors fail a lot in most other agent products, not on ours. So solid connector infrastructure. And of course, our search advantages, being able to spin up browsers on the sandbox. These are all things outside the model…
Yeah. So let's actually talk about the progression of your products, because I find it very interesting. And I'm sure I'm going to mess up the timeline, so just cut in when I get it wrong. But you obviously have your answers engine, and then I believe last year you built Comet, which is a browser.…
we really recognize that you need to be where users are, which is why we built an browsing agent, because most people spend a lot of time on their browser. Sure. But what we realized is people like this conversational interface using things from their phone, being able to delegate their tasks from…
Okay, let's address that. Yeah, well, so. It was this time a year ago where there was a vote. It wasn't to fail. They said they wanted to short you guys.
More from Newcomer Pod
