Parallel’s Parag Agrawal: Building a New Web for AI Agents
- Source
- youtube.com
- Author
- Sequoia Capital
- Date
Context
Parag Agrawal, the former Twitter CEO who now runs Parallel Web Systems, says search for AI agents should differ from search for people because agents do not generate human click data. Search engines normally learn from human ratings and clicks. In this Sequoia Capital interview Agrawal calls human click data a bug for search and says models can now supply such ratings more cheaply, which is the bet behind Parallel's ranking.
Agrawal also explains how a small company could compete with Google-scale indexes. Parallel launched first as a search agent that crawls pages after a query arrives, which suits deep-research jobs where a minute's wait is acceptable, and built its index up as customers used it. Early users were replacing outsourced human data gathering in insurance, sales and finance. Agrawal says a newly shipped product now selects results in 200 milliseconds, down from a three-second budget. These are the company's claims, with no benchmarks shown. Agrawal discusses accuracy, and cost tradeoffs for agent search in a separate interview.
- Agent — A system in which a model repeatedly chooses operations, a harness carries them out, and the results return to the model to inform its next choice.
- Latency — The time between asking for a result and having it, including the model's generation time, tool calls, retries and any queueing.
- Model — A computational system that uses learned patterns to produce an output from information supplied to it.
videoBringing Continual Learning to Every Company: Trajectory’s Ronak Malde
videoCrusoe CEO Chase Lochmiller: Turning Stranded Energy Into Intelligence
videoDavid Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
videoCracking the Code on Offensive Security With AI ft XBOW CEO and GitHub Copilot Creator Oege de Moor
Checking sign-in…
Loading comments…


