🛠 Tech Stack

💼 About This Role

You'll join an applied AI research team building code agents for real-time interactive content. You'll design agentic systems and post-training pipelines, with a focus on multi-step coding evaluation and secure execution. This role offers the chance to shape core technology from research to production.

🎯 What You'll Do

• Design agentic systems with tool catalogs and function calling
• Build evaluation suites for multi-step coding tasks
• Develop post-training pipelines using RL and preference optimization
• Implement secure sandboxed execution environments

📋 Requirements

• 3+ years experience in machine learning or AI engineering
• Strong background in reinforcement learning or RLHF
• Experience shipping agentic systems that pass real repository tests
• Proficiency in Python and deep learning frameworks

✨ Nice to Have

• Published papers in agentic systems or code generation
• Contributions to agent frameworks like LangGraph or AutoGen
• Experience building datasets from execution traces

🎁 Benefits & Perks

• 🩺 Health insurance
• 💰 Equity package
• 🍽️ Daily meals
• 🏢 On-site collaboration