AI Infrastructure Engineer — LLM Reliability & Observability (Contract, Project-Based)

Descrição da Vaga

Requisitos

• Hands-on experience shipping AI/LLM systems in production (not just experimentation)
• Familiarity with observability concepts — traces, spans, latency, reliability metrics
• SRE or DevOps background is a strong plus
• Comfortable working async with a small founding team
• Work on genuinely interesting reliability infrastructure — not CRUD apps
• Founding-team access — you'll work directly with the founder, not a PM layer
• Flexible between engagements — no pressure to be available when there's no active work
• If you've shipped AI systems in production and want to work on the infrastructure layer that makes them reliable — apply and tell us about the most complex AI system you've deployed

Responsabilidades

• Think: regression detection, deployment safety gates, runtime guardrails, and incident investigation for AI systems
• We're building a bench of engineers to contract on client implementation engagements as they close
• This is **not a permanent role** — it's project-based work that starts when a client engagement is confirmed
• *What the work looks like:**
• Deploying the Reliai platform into client production environments (Python SDK, Node SDK)
• Configuring trace ingestion, guardrails, and deployment safety gates for each client's AI stack
• Integrating with client CI/CD pipelines and model deployment workflows
• Working with OpenAI, Anthropic, LangChain, and LlamaIndex integrations
• Duration: 6–10 weeks per engagement, potentially recurring

Benefícios