SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Alan is a prevention-focused health insurance company serving 1M+ members across Europe and Canada. The company integrates insurance, prevention, and care delivery into a unified platform, with €800M+ ARR and 1000+ employees.
You will join the product team building AI agents that answer member questions about health insurance—reimbursements, coverage, claims, account changes—instantly and at scale. Alan's agents already fully resolve 50%+ of member inquiries. Your mission is to push that resolution rate higher, improve answer quality and member trust, and integrate these agents with other AI systems like Mo (Alan's health companion).
Key responsibilities:
- Design and ship LLM-powered agents that understand member requests, reason over contracts/claims/history, and take real actions on their behalf; build tools, retrieval pipelines, and context systems to provide agents with timely information
- Enable agent interoperability across Alan's AI ecosystem, handling conversation handoffs, context sharing, and capability chaining for seamless member experiences
- Expand automated resolution to new channels, starting with AI phone support for real-time voice conversations matching written quality standards
- Own evaluation: build datasets, automated evals, and quality monitoring to validate agent correctness before member exposure
- Measure and iterate on resolution rate, answer quality, and member satisfaction in production; optimize prompts, models, and architecture
- Partner with care experts and product teams to identify new topics for AI automation
This role offers direct impact—every percentage point of automated resolution means instant member help and drives Alan's growth and profitability. You'll ship production AI agents handling hundreds of thousands of real conversations monthly in a regulated, high-stakes healthcare domain where accuracy matters. You'll tackle frontier problems: agentic workflows, tool use, evaluation, safety, guardrails, and cost/latency trade-offs in LLM systems. You'll have ownership and growth opportunities, influencing product direction and mentoring other engineers.
Tech stack: React, Python/Flask, PostgreSQL in a monorepo. Daily deployments; distributed ownership model where builders own what they ship.
Requirements:
- 5+ years building scalable software systems
- Strong backend experience (Python preferred)
- Hands-on experience building and shipping LLM-powered products to production (agents, tool calling, RAG, prompt engineering, evaluation)
- Evaluation mindset: ability to measure AI system quality and drive improvements
- Passion for turning complex problems into elegant solutions
- Curiosity about the "why" behind your work, not just the "how"
- Pragmatic approach to fast-moving AI landscape
- Healthcare technology interest and understanding that insurance is about access to care
- Legal eligibility to work in France, Belgium, or Spain
