Pranav Putta
Pranav Putta is the co-founder and chief technology officer of the AI start-up PlatoPlatoPlato develops simulated work environments for training and evaluating AI agents. It builds working environments around the software, data and workflows of real jobs and publishes benchmarks for assessing model capabilities.Open the full entry and a researcher on autonomous AI agents. He is the first author of Agent Q, an agent research paper from MultiOn.
Contents4 sections
Key facts
August 2024Agent Q
In August 2024 he was first author of the agent paper Agent Q, with his affiliation listed as The AGI Company (MultiOn); co-authors included Stanford’s Chelsea FinnChelsea FinnChelsea Finn is a computer scientist and robot-learning researcher who is an assistant professor of computer science and electrical engineering at Stanford University and a co-founder of the robotics AI company Physical Intelligence. Known for work on meta-learning and robot learning, she won the 2018 Doctoral Dissertation Award of the Association for Computing Machinery and was named to the TIME100 AI list in 2026.Open the full entry and Rafael Rafailov.[1]
The method combines guided Monte Carlo Tree Search, self-critique and an off-policy variant of DPO; in the paper’s real-world booking experiments, Llama-3 70B’s zero-shot success rate rose from 18.6% to 81.7%.[1]
That month MultiOn opened early access to Agent Q through a waitlist.[2]
2025The REAL Benchmark and Plato
In April 2025 he co-authored “REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websites,” which introduced a benchmark for evaluating autonomous agents on multi-turn tasks in deterministic simulations of real websites; his affiliation was listed as Plato.[3]
He co-founded Plato with Robert FarlowRobert FarlowRobert Farlow is the co-founder and chief executive of Plato, a San Francisco AI start-up founded in 2025 that builds simulated versions of real-world work software for training and evaluating AI agents.Open the full entry and serves as its chief technology officer.[4]
Plato builds simulated environments for training and evaluating browser agents under the tagline “Experience, Beyond Imitation.”[5]
October 2026Series A
In October 2026 Plato announced a $16 million Series A. “We build evaluations that reflect what the job actually requires,” he said.[4]
Sources
- Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents (opens in a new window)
- Multion AI Agent Q is alive! (opens in a new window)
- REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websites (opens in a new window)
- Plato Raises $16 Million Series A to Build Simulated Workplaces for AI Agents (opens in a new window)
- Plato | Simulated Worlds for Agents (opens in a new window)