P

Plato

Delivers reproducible AI training environments

Member of Technical Staff - Research Engineer

AI ResearchConfirmed live in the last 24 hours
Full-timeSan Francisco, CA, USA$200k - $400k /yrIn-person

About the Job

Requirements

  • Strong implementation ability and the ability to turn ambiguous research ideas into working systems.
  • Experience with reinforcement learning, large language model agents, computer-use agents, evaluations, post-training, synthetic data, simulation, or model behavior analysis.
  • Care about whether a task is grounded, difficult, resistant to reward hacking, and capable of producing actual learning signal.
  • Comfort interpreting ambiguous model behavior and negative results.
  • Enjoy building continuous research loops rather than static benchmark artifacts.

Responsibilities

  • Discover model failure modes from real-world traces, agent telemetry, targeted researcher hypotheses, and customer workflows.
  • Generate realistic curricula grounded in actual workflows rather than toy synthetic benchmarks.
  • Benchmark candidate tasks against frontier computer-use and agent models using pass rates, rollouts, and behavioral traces as difficulty signals.
  • Build hill-climbing loops that mutate, filter, and rescore tasks until they surface high-signal targets.
  • Study reward hackability, distribution mismatch, task realism, long-horizon failures, and transfer from simulation to deployed agents.
  • Turn research prototypes into reliable internal systems for continuous curriculum generation.
  • Design experiments, build task generation systems, run evaluations, inspect model failures, and develop methods for mining tasks that are just out of reach of today's agents.
  • Consume real-world trajectories or researcher hypotheses, materialize realistic data, propose candidate tasks, benchmark those tasks against frontier computer-use and agent models, and hill-climb until finding failures that produce useful learning signal.
  • Own the full loop of hypothesis, implementation, evaluation, analysis, iteration, and productionalization.

About the Company

Plato

Delivers reproducible AI training environments

Plato creates datasets and simulated environments to train and evaluate AI web agents. It produces static, reproducible environments from human demonstrations, providing labs and agent developers a controlled way to benchmark and reinforce-behavior learning on real-world web tasks. Unlike other providers, Plato focuses on supplying the data and virtual “gyms” that power modern AI progress in reinforcement learning. Its goal is to help advance AI by making reliable, testable environments and benchmarks for web-based tasks.

Company size
11-50
Company stage
Series A
Total funding
$16M
Headquarters
San Francisco, CA
Founded
2024

Recently Posted Jobs

Explore more opportunities at Plato

See more jobs