Applied Research - RL & Agents
Core
Designing and implementing next-generation AI agents, reinforcement learning (RL) methods, and post-training infrastructure to align large models with real-world workloads.
Role type
Senior Applied Research Engineer (RL & Agents)
Builds
Open superintelligence stack, agent frameworks, RL training services, and evaluation harnesses
Domain
Frontier AI, Reinforcement Learning, Agent Systems, Infrastructure
Deliverable
production ML models | product features | research
Required skills
Machine learning engineering, Reinforcement Learning (RL), Post-training methods, Agent frameworks, Distributed training/inference, Technical writing, Research contributions
Preferred skills
Web programming (React, TypeScript), LLM evaluations, Synthetic data generation, Container orchestration (Docker, Kubernetes)
Technologies
DSPy, LangGraph, MCP, Stagehand, vLLM, sglang, Accelerate, Ray, Torch, Prometheus, Grafana
Responsibilities
Design and iterate on AI agents for workflow automation and decision-making; Develop RL and post-training methods (RLHF, RLVR, GRPO) for model alignment; Build evaluation harnesses to measure reasoning and agentic behavior; Architect distributed training/inference pipelines for scalability; Prototype multi-agent and memory-augmented systems; Translate research objectives into technical requirements for product teams.
Seniority
Senior, hands-on IC
