Exowatt is revolutionizing the energy landscape for the AI era with our groundbreaking P3 system that captures solar energy, stores it as heat, and generates electricity on demand. Founded in 2023 and backed by leading investors including Andreessen Horowitz, Sam Altman, and Felicis, we're committed to providing clean, modular, and scalable power that meets the rapidly growing demands of AI infrastructure. Our mission is to make sustainable renewable energy always available and almost free, enabling technological advancement while protecting our planet.
Work as one of a two-person intern team extending Exowatt's internal AI agent platform: building, deploying, and hardening agents that the whole company uses, and pushing into LLM fine-tuning and agent-architecture work tied to our AWS inference fellowship.
This position is based at our Austin facility.
Key responsibilities
Build and ship new agent capabilities on the internal platform, deployed company-wide.
Contribute to agent-platform architecture (tool/skill surfaces, runtime, deployment patterns).
Work on LLM fine-tuning / custom-inference experiments — e.g. cheaper/faster models that hold frontier-level reliability for monitoring and ops workflows.
During onboarding, read and improve existing internal apps (real platform improvements, not throwaway exercises).
Help run weekly internal AI Office Hours teaching the broader company to use our AI agent tools.
Required qualifications
Pursuing a CS / software-engineering (or closely related) degree.
Strong programming fundamentals; comfortable in Python.
Working familiarity with LLMs / AI APIs and a genuine eagerness to build with them.
Comfortable with Git and collaborative development.
Preferred / nice-to-have
Experience with agent frameworks, RAG, MCP, or prompt/tool design.
Exposure to model fine-tuning or inference optimization.
Cloud experience (AWS).
Self-directed; thrives with real ownership in a fast-moving environment.
