Site Reliability Engineer

Arena Intelligence is looking for an engineer to build the core infrastructure that sits beneath our online evaluation systems - the AI gateways, automated arena runtimes, and serving layers that make real-world model evaluation possible at scale.

This is a critical part of the Arena Service. Arenas are live, online systems: they route traffic across frontier models from many providers, handle bursty and unpredictable load, need to fail gracefully when upstream models do, and have to remain fair and consistent under all of it. We exist to build foundational infrastructure for our users that scales, is reliable, and makes the complexities of operating this infrastructure at scale disappear. We need a practitioner who’s shipped this kind of infrastructure before and knows where the sharp edges are.

You’ll be an early member of our infrastructure team, working closely with researchers, engineers, and product leadership. The work is zero-to-one in places and scale-it-up in others. We move fast and stay rigorous., * Build and orchestrate API-based products from the ground up. Design and implement low-latency, high-reliability APIs for leaderboards, models, and arenas.

• Ship enterprise-grade infrastructure. Build the systems enterprise customers expect: rate limiting, authentication, usage metering, cost attribution, audit logging, and SOC 2 compliance.
• Build deep observability. Instrument infrastructure with distributed tracing, latency breakdowns, token-level usage tracking, and real-time dashboards so customers (and we) can see exactly what’s happening.
• Build AI-centered products. Integrate with our core evaluation platform, Arena data, and customer-specific benchmarks. Collaborate with the research team to turn novel ideas into full-featured products.

• 6+ years of backend engineering experience, with meaningful time spent on distributed systems, infrastructure, or developer-facing platforms.
• Strong proficiency in Go and/or Rust, with hands-on experience building high-throughput APIs or proxy/gateway systems.
• Experience with LLM provider APIs (OpenAI, Anthropic, Google, etc.) and a working understanding of the challenges: streaming, token management, rate limits, model-specific quirks.
• Solid cloud infrastructure skills - you’re comfortable with AWS or GCP, Kubernetes, Terraform, and database systems like Postgres and Redis.
• A product-oriented mindset. You think about the developer experience of your APIs, not just the implementation. You ask “why” before “how.”
• Comfort with ambiguity. We’re a startup. Scope is fluid, context shifts, and you’ll wear many hats. That should sound exciting, not stressful., * Experience building API gateways, proxies, or developer tools (Bifrost, Kong, Envoy, Tyk, or custom).
• Background in ML infrastructure, model serving, or evaluation frameworks.
• Experience building enterprise-ready features: SSO, RBAC, audit logs, multi-tenancy.
• Experience building billing infrastructure around systems like Stripe, Metronome and Orb
• Familiarity with the modern AI infra stack (vLLM, LiteLLM, LangChain, etc.).

• We offer competitive compensation and equity aligned to the markets where our team members are based. The base salary range will depend on the candidate’s permanent work location.
• Comprehensive health and wellness benefits, including medical, dental, vision, and additional support programs.
• The opportunity to work on cutting-edge AI with a small, mission-driven team
• A culture that values transparency, trust, and community impact

Come help build the space where anyone can explore and help shape the future of AI.

About Arena Intelligence

Arena is the platform for evaluating how AI models perform in the real world. Founded by researchers from UC Berkeley’s SkyLab, we’re on a mission to measure and advance the frontier of AI for real-world use, and to build the foundation for everyone to understand, shape, and benefit from it.

Tens of millions of people use Arena each month to evaluate how frontier systems handle the work they actually do. The preferences they share power the most transparent, rigorous, and human-centered evaluations in AI. Leading AI labs, enterprises, and independent researchers rely on our work and open datasets to understand how models behave in real workflows: agentic coding, creative generation, professional productivity, and beyond. We go beyond leaderboards and decompose what human experience reveals about AI, so models advance toward the work people actually do.

We’re a team of researchers, academics, builders, and creatives from UC Berkeley, Google, Stanford, and DeepMind. We seek truth, move fast, and value craftsmanship, curiosity, and impact over hierarchy. We’re building a company where thoughtful, curious people from all backgrounds can do their best work together, in an office culture that radiates excellence, energy, and focus.