Founding Member of Technical Staff

About the role

The role involves building software to significantly enhance the quality and efficiency of Reinforcement Learning (RL) environment creation, crafting domain-specific verifiers for various financial services verticals, and researching and developing advanced RL environments. The candidate will also establish the engineering culture and practices from the ground up.

Responsibilities

• Building software to 10-100x the quality and throughput of RL environment creation
• Create platform analytics to track data on RL environment costs, hours, bottlenecks, and subject matter expert management
• Craft domain-specific verifiers for financial services verticals and beyond (e.g., verifiable rewards for powerpoint building, excel modeling, quantitative trading)
• Research and develop next-generation RL environments that are increasingly realistic, long-horizon, and challenging for frontier models
• Establish engineering culture and practices from the ground up

Requirements

• Familiarity with building evaluation, benchmark, and RL for AI agents
• Smart usage of AI coding tools like Claude Code, Cursor, Windsurf, etc., while also implementing best practice testing and code quality attention to detail
• Startup Speed: Iterate quickly, ask questions quickly, respond to messages quickly, and make mistakes quickly
• Product/user ownership, ability to prioritize between a large roadmap of features
• Client-facing capabilities - comfortable talking to users/customers

Nice to have

• Research backgrounds (training models, publishing papers)
• Experience with early-stage startups
• Previous experience working with data companies (ex. Surge, Mercor, Handshake)
• Side projects involving AI Agents/RL

Benefits

• Fully covered healthcare coverage for yourself and dependents (Medical, Dental, Vision)
• Relocation Stipend ($10k)
• 30-minute meet and greet with Jerry Wu (Founder / CEO)
• 30-minute meet and greet with Wyatt Marshall (Founder / CTO)
• 30-minute meet and greet with Will Bryan (Founding Member Technical Staff)
• On-site (2 days) or remote (7 days) paid work trial