The client, a pioneering force in artificial intelligence, is looking to build AI agents that power every part of the company from Marketing and GTM to Recruiting, Support, Internal Tools, and their Core Product. They aim to transform LLM systems into dependable production experiences and accelerate the mission to make intelligent, autonomous software a reality both internally and for their customers.
Requirements
- Experienced software engineer with a strong track record shipping AI or ML-powered applications (typically 3+ years, including at least 1 year building LLM systems in production).
- Hands-on experience implementing evaluation and monitoring systems for agents or workflows.
- Deep understanding of the components that make up an AI system: prompting, retrieval, orchestration, inference APIs, and model selection across modalities.
- Strong coding skills in Python or TypeScript (ideally both).
Responsibilities
- Design, implement, and deploy end-to-end AI workflows and agents that solve real problems across multiple business domains.
- Develop and iterate on agent architectures, evaluation pipelines, and performance frameworks to ensure reliability and measurable outcomes.
- Translate emerging AI research and tooling into practical, production-ready solutions.
- Communicate technical decisions, trade-offs, and insights clearly to both technical and non-technical stakeholders.
- Collaborate cross-functionally embedding with teams like Marketing, GTM, Recruiting, or Product to identify opportunities for agent-driven automation and measurable business impact.
- Contribute to the open-source ecosystem, including open source components, documentation, and shared tools.
Other
- Thrives in a fast-moving, ambiguous startup environment; enjoys identifying the highest-impact problems and driving them to completion.
- Naturally curious and motivated to learn new tools, frameworks, and approaches in applied AI.
- Excellent communicator who can simplify complex technical ideas for diverse audiences.