SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Fever is a global tech platform for culture and live entertainment, reaching 300+ million people monthly across 55+ countries. The company partners with major brands like Netflix and F.C. Barcelona to democratize access to cultural experiences.
You'll join the AI & DevEx team, which focuses on raising the performance of Fever's entire Product Engineering organization through AI-powered tools and infrastructure. This is fundamentally a backend engineering role with AI expertise layered on top.
Key responsibilities include:
**Agentic Infrastructure**: Design and build the skills, sub-agents, guardrails, and verification systems that ensure agents produce high-quality, trustworthy output across the company.
**Knowledge Base**: Develop the shared knowledge substrate that agents read from and write into, scaling from team-level to company-wide.
**Delivery Substrate**: Own cloud dev environments, CI runtime, and autonomous-delivery pipelines that transform requests into environments, agent runs, and reviewable pull requests. You're accountable for speed and reliability.
**Squad-Level AI**: Move beyond coding agents to identify where AI removes friction in planning, coordination, and communication across engineering squads.
**Measurement & Evals**: Build evaluation and metrics systems (baselines, before/after comparisons, DORA metrics, lead time, adoption tracking) to ensure decisions are signal-driven, not assumption-based.
**Inference Economics**: Optimize AI workflows for cost and speed through model routing, prompt caching, and self-hosted vs. frontier model trade-offs.
**Engineering Excellence**: Set the bar for testing, design patterns, CI/CD, and observability. Everything ships as production software.
You'll need 5+ years building and operating production software at scale with end-to-end ownership, strong backend fundamentals (APIs, SQL, distributed systems, event-driven patterns), and full production ownership (CI/CD, observability, infrastructure). On the AI side, you need 1–2+ years shipping LLM applications to production (not prototypes), hands-on experience with context engineering, RAG, multi-agent orchestration, and evals. You should have personally authored skills, sub-agents, evals, or verification harnesses and can demonstrate real work. You think in measurements and kill experiments that don't move the needle.
Bonus: public work on agent harnesses or AI evaluation (open source, talks, papers), experience building internal developer platforms, and comfort operating with ambiguity at startup speed.
