🛠 Tech Stack

💼 About This Role

You'll own and improve core action models end-to-end for Composite's proactive automation platform, working at the intersection of LLM inference, browser understanding, and low-latency systems. Your work will ship directly to users, improving the work lives of millions. You'll tackle unique ML challenges from running AI through existing browsers via a Chrome extension.

🎯 What You'll Do

• Improve accuracy and latency of core models across web applications
• Design and optimize LLM inference pipelines with caching and streaming
• Build evaluation frameworks and data pipelines for model quality
• Experiment with retrieval-augmented approaches using vector databases

📋 Requirements

• Strong ML fundamentals with production deployment experience
• Obsessive about latency and optimizing inference pipelines
• Deep care about data quality and building tooling to ensure it
• Experience with LLMs, transformer architectures, or sequence prediction

✨ Nice to Have

• Experience with browser automation or Chrome extensions
• Familiarity with accessibility tree or DOM parsing
• Background in RL or online learning from user interaction data

🎁 Benefits & Perks

• 💰 Competitive equity package
• 🏥 Health is #1 value (sleep, exercise, mental health)
• 🍔 Balanced diet encouraged
• 🏋️ Exercise frequently culture

📨 Hiring Process

Send resume and a brief note about why you are a good fit.