Job Description

About the Role

You'll build the learning system for autonomous trading agents — enabling the fleet to learn from its own decisions and continuously improve.

This is a production ML role (not research) focused on reinforcement learning and closed-loop systems.

Key Responsibilities

Learning System & RL Loop

• 
  Build feedback loop from trade outcomes strategy improvement
• 
  Develop evaluation frameworks for signals and performance
• 
  Automate strategy generation and backtesting
• 
  Detect market regime shifts and adapt strategies
• 
  Implement performance attribution systems
• 
  Manage fleet coordination (risk, capital allocation)
• 
  Build telemetry/data infrastructure

Model & Inference

• 
  Define and implement model hosting strategy
• 
  Build domain-specific training pipelines
• 
  Optimize inference for real-time trading agents
• 
  Capture full decision telemetry across agents

Requirements

• Closed-loop production ML systems (non-negotiable)
• 
  RL / online learning experience
• 
  Full-stack ML ownership (data model production)
• 
  Strong Python + backend experience (Go/TS helpful)
• 
  High-stakes system experience (preferred)

Bonus Skills

• 
  Financial ML / trading systems
• 
  LLM fine-tuning & serving
• 
  Multi-agent systems
• 
  DeFi / onchain experience
• 
  Experience in sequential decision systems (robotics, game AI, etc.)