Vmax
About
Vmax builds self-play infrastructure that generates new RL tasks and environments from proprietary data and evaluations. Its research focuses on task generation, RLVR, long-horizon learning, and automatically improving agents.
What They Offer
Commercial or operational services this company provides.
Agent evaluations
Rubrics / verifiers
Custom RL environments
Managed infrastructure
Products & Public Artifacts
Campaign
Environment · Access by request
Self-play task-generation infrastructure that transforms proprietary data/evals into environments and refines agents.
Visit artifact →unix-ctf
Environment · Public
Procedural capture-the-flag environments for training and evaluating Unix competence in language-model shell agents.
Visit artifact →Technical Capabilities
Code / Terminal ExecutionSandboxed ExecutionLong-Horizon TasksProgrammatic Verifiers
Only publicly documented capabilities are shown. Absence does not imply the capability is unavailable.
Focus Areas
RL EnvironmentsRLHF / Post-trainingEvaluations
Domains
Software Engineering
Leadership
Matthew Sargent
Co-founder
Augustine Mavor-Parker
Co-founder and CTO