Role Overview
We’re hiring an early AI engineer to build systems that make high-stakes decisions, learn from real-world outcomes, and improve continuously.
The product is already live with strong data flow and user activity. Your job is to build the intelligence layer on top — turning that data into systems that get better over time.
You’ll own AI end-to-end: defining what to build, designing learning systems, and shipping production code. No predefined roadmap — you’ll work directly with the founder to shape the direction.
Focus Areas
- Closed-loop learning systems (RL, recommendation, ranking)
- Autonomous agents and workflow automation
- Model infrastructure, inference, and data pipelines
What You’ll Do
- Build systems where decisions → outcomes → automatic improvement
- Develop recommendation / decisioning systems that learn from real-world feedback
- Design evaluation frameworks and simulations
- Fine-tune and deploy models using proprietary data
- Build agents that automate complex workflows
- Own everything from data → model → production
Requirements
- Proven experience building closed-loop systems in production (non-negotiable)
- Background in RL, recommendation systems, or adaptive learning systems
- Strong end-to-end ownership (data, models, deployment)
- Proficient in Python + modern AI stack (LLMs, embeddings, etc.)
- Comfortable in fast-moving, ambiguous environments
Ideal Background
- Built AI systems that improved from real user or system feedback
- Early-stage or product-focused experience
- Experience with sequential decision-making (e.g., marketplaces, robotics, optimization)
