Research Engineer - Post-Training for Agentic Coding
Core
Developing and implementing advanced products that enable customers to post-train large language models for agentic coding practices, ensuring generated code meets enterprise standards.
Role type
Senior Research Engineer (LLM Post-Training & Agentic Systems)
Builds
Enterprise-grade coding agents and models for AI-assisted software development
Domain
AI Software Development / Large Language Models / Code Verification
Deliverable
production ML models
Required skills
Reinforcement learning from verifiable rewards, GRPO techniques, offline/semi-online reinforcement learning, parameter efficient fine-tuning, supervised fine-tuning, safety alignment, large-scale data processing, cloud infrastructure management
Preferred skills
Python, Rust, C#, C++, JS/TS, Java, AWS, Microsoft Foundry, Databricks
Technologies
Python, Rust, C#, C++, JS/TS, Java, AWS, Microsoft Foundry, Databricks
Responsibilities
Design hypotheses and experiments to iterate proofs-of-concept into cutting-edge products, contribute to cross-disciplinary discussions on next-generation coding model post-training, explain complex technical concepts to technical and non-technical audiences
Seniority
Senior, hands-on IC
