Senior/Staff AI Software Engineer

Hey there!

We’re Fever , the world’s leading tech platform for culture and live entertainment.

Our mission? To democratize access to culture and entertainment. With our proprietary cutting-edge technology and data-driven approach, we’re revolutionizing the way people engage with live entertainment.

Every month, our platform inspires over 300 million people in +55 countries (and counting) to discover unforgettable experiences while also empowering event creators with our data and technology, helping them scale, innovate, and enhance their events to reach new audiences.

Our results? We’ve teamed up with major industry leaders like Netflix, F.C. Barcelona, and Primavera Sound, presented international award-winning experiences, and are backed by several leading global investors! Impressive, right?

To achieve our mission, we are looking for bar-raisers with a hands-on mindset who are eager to help shape the future of entertainment!

Ready to be part of the experience?

Now, let’s discuss this role and what you will do to help achieve Fever’s mission.

About the role

We’re building an agentic software-delivery system: AI agents that write, test, and review code, with the goal of shipping that code to production without human intervention. You’ll help build it, and you’ll teach the rest of Fever’s engineers to use it.

You’ll join AI & DevEx, the team whose mission is to maximize the performance of every squad at Fever through AI. A squad is far more than code. It’s product management, product design, and engineering, plus everything around them: CI/CD, observability, infrastructure, data, and the way people communicate and coordinate. Almost everywhere in that chain, from a first idea to a shipped, validated feature, there’s something AI can make better. We find the highest-leverage points and improve them, always backed by measurement.

We’re not starting from scratch. AI agents are already in active production use across our engineering workflows, and we’ve built evaluation infrastructure to measure and improve their output. This role is about taking us from advanced to best-in-class.

This is a software engineering role at its core, not a data science or ML research position. We’re looking for an outstanding engineer with strong fundamentals who has built real AI and agent expertise on top. You’ll report to the Head of Engineering, AI & DevEx, and you’ll stay hands-on no matter your level. Depending on how your profile calibrates during the process, you could lead a small team of AI engineers, or drive key initiatives as a senior individual contributor with room to grow into leadership.

You’ll work on a stack built on:

• Python/Django and FastAPI

• Angular

• Celery, RabbitMQ and Kafka

• AWS

• Hexagonal architecture

• CI/CD on Jenkins

What you'll work on

• The agentic harness. Build the skills, sub-agents, guardrails, hooks, and verification that make agents produce shippable output, not suggestive output, across the whole software lifecycle. Turn non-deterministic agents into reliable systems through staged workflows, adversarial review, and automated quality gates.

• A knowledge base that compounds. Build the shared knowledge substrate that agents read from and write into, including architecture conventions, domain knowledge, and past decisions. Grow it from team level into something that compounds across the entire Fever organization.

• The delivery substrate and its autonomous rails. Own cloud dev environments, CI runtime, ephemeral environments on demand, and the rails that turn a request into an environment, an agent run, and a pull request with reviewable evidence. You own how fast it is and how reliably it works.

• Evaluation and measurement. Build the measurement substrate that lets every change land on signal, not vibes. That means eval sets built from real tasks in our own codebase, naked baselines, and “ship only on a measured improvement.” The metrics that matter are:

• lead time from idea to production

• DORA metrics

• cost per shipped feature

• shippable-output rate

• human intervention rate

• real adoption

• Inference economics. Make AI-driven workflows fast and cost-bounded. Decide on model routing, prompt caching, and tool-list pruning, and know when self-hosted open-source models beat frontier models on quality and cost.

• Adoption and enablement. Give internal talks, workshops, and mentorship sessions. Work directly with the squads we serve so the tools actually get used, and look beyond coding to how PMs, designers, and engineering leads plan, coordinate, and communicate.

• Engineering excellence. Testing, design patterns, CI/CD, and observability. Everything here ships as production software, so the people who build it set the bar for what good code looks like.

• Technical leadership. Drive key initiatives end to end and mentor other engineers. Depending on your level, this can include leading and growing a small team of AI engineers while staying close to the code.

About you

• Production experience: 5+ years building and operating production software at meaningful scale, such as distributed services, high traffic, or systems many teams depend on. You’ve owned systems end to end and can explain the trade-offs behind your decisions, not just what you built.

• Software design: Strong fundamentals: clean architecture, design patterns, clear boundaries. You move easily between the big picture and the detail of the code.

• Backend depth: RESTful APIs, relational databases and SQL, and the async and event-driven patterns (queues, messaging, idempotency) behind high-traffic services.

• Quality is your job, not QA’s: You bring test-driven development, a test strategy across unit, integration, and end-to-end, static analysis, and CI quality gates.

• Production ownership: CI/CD, observability and monitoring, and accountability for how your code behaves once it’s live. “Done” means healthy in production, not merged.

• LLM applications in production: At least the last 1–2 years deep in LLM applications shipped to production, not just prototyped. You have a working command of context engineering, RAG, multi-agent orchestration, and evals.

• You build with agents; you don’t just use them: You use a coding agent every day on production work, and you’ve engineered it with your own skills, sub-agents, hooks, MCP integrations, and knowledge bases. You can walk us through something real you built and explain why you built it that way.

• You think in measurements: Baselines, before/after comparisons, accounting for non-determinism, and killing experiments that don’t move the needle.

• AI observability: You understand that AI systems often degrade silently, and you know how to monitor for it.

• Outstanding communicator in English: You explain complex technical topics clearly to engineers at every level, through mentoring, tech talks, workshops, or technical writing.

• Builder at heart: You want to grow your impact beyond your own code by mentoring others, leading initiatives, and, at Staff level, leading a team while still writing code. You’re as comfortable in a room with executives as in a pairing session.

• Product mindset: Internal tools have users too. You think about adoption, prioritization, and impact you can measure.

Bonus Points

• Public artefacts on agent harnesses, AI evaluation, developer experience, or AI-era engineering tooling: open source, talks, blog posts, papers.

• Experience building internal tools or developer platforms where other engineers are your users.

• Experience leading or mentoring a team of engineers.

• Comfort operating with ambiguity at startup speed: small teams, no playbook, decisions you can’t outsource.

Benefits & Perks

• Attractive compensation package consisting of base salary and the potential to earn a significant bonus for top performance.

• Stock options.

• Opportunity to have a real impact in a high-growth global category leader

…