9 days ago
Remote, EMEASenior
Responsibilities
- Design and build reusable agentic infrastructure for memory, context management, tool-calling, orchestration, and feedback loops.
- Develop evaluation, tracing, debugging, logging, and observability systems for production AI agents.
- Create workflows using coding agents and specialized subagents to explore repositories, implement changes, review architecture, and validate outputs.
- Design systems coordinating agents, tools, deterministic software, and human intervention.
- Translate complex operational workflows into reliable production automation with operational and product teams.
- Help define architectural patterns, tooling, and engineering standards for agentic systems as the company scales.
Requirements
- Strong software engineering background with experience independently delivering complex systems from concept through production.
- Hands-on experience building AI or agentic systems that have operated in production.
- Experience with reliability, latency, context management, failure modes, evaluation, and observability for real-world AI systems.
- Strong understanding of agents, tool use, structured outputs, context management, orchestration, evaluation, and feedback loops.
- Advanced practical use of modern LLMs and reliable workflows around probabilistic systems.
- Experience using coding agents, agentic development workflows, and specialized subagents in engineering work.
- Strong architectural judgment regarding LLMs, deterministic software, human intervention, and combined approaches.
- Experience with TypeScript and Node.js, or strong Python-based AI systems experience with willingness to work in a TypeScript-first environment.
- Strong product judgment, autonomy, ownership, communication skills, English fluency, and ability to work under ambiguity.
- Startup mentality, resilience, adaptability, and comfort in a fast-paced environment.
Benefits
- Fully remote work across the UK, Ireland, and European time zones.
- Competitive salary and potential equity options based on performance.
Tech Stack
Categories
