Sr. Software Engineer, AI

What you’ll do

You’ll own AI infrastructure that serves every team in the company — we expect the work you build in your first year to save thousands of hours annually via 50+ new AI agents. You’ll embed with internal teams, find the highest‑leverage automation opportunities, and own them end‑to‑end: discovery, simplification, build, deployment, and adoption. You’ll scope a problem with a non‑technical stakeholder in the morning and ship production infrastructure in the afternoon.

You measure your work in hours unlocked and cycle time reduced — not stories closed.

• Design and build multi‑step agentic workflows in Python and Type Script — planning loops, tool dispatch, error recovery, and explicit human‑in‑the‑loop checkpoints for high‑stakes decisions
• Develop production LLM applications on Anthropic and OpenAI SDKs, including prompt engineering, structured outputs, tool/function calling, prompt caching, and batch processing
• Build and maintain RAG pipelines — embedding generation, vector/hybrid search, knowledge base ingestion — and apply judgment about when retrieval actually helps versus adds noise
• Own eval discipline end‑to‑end: define offline eval sets, run A/B experiments on model changes, build regression suites, and articulate “good enough” exit criteria using Lang Smith, Braintrust, or equivalent
• Drive cost and latency optimization — token budgets, model tier selection (Haiku / Sonnet / Opus and GPT equivalents), and caching strategies that hold up at scale
• Build MCP servers and function‑calling connectors that give agents reliable, schema‑governed access to internal tools, APIs, and data sources — Jira, CRM, Slack, internal services, and more
• Implement and maintain production integrations using REST, Graph
  
  QL, webhooks, and event‑driven patterns (queues, Pub/Sub) with proper idempotency, retry logic, and backfill support
• Wire up OAuth/SAML authentication flows (Okta in particular) for secure agent‑to‑service access across internal and third‑party systems
• Own cloud infrastructure for AI workloads on GCP using Terraform, GKE/Cloud Run, and secrets management — with logging, metrics, and alerting from day one
• Build data pipelines that feed AI systems: strong SQL, Athena/Big Query‑class warehouses, ETL/ELT, schema design, and data‑quality monitoring
• Partner with internal teams across Engineering, Operations, Customer Support, Data, and Finance to identify where agentic automation can have the highest leverage — then build it
• Create reusable libraries, SDKs, and internal tooling so teams can extend AI capabilities without starting from scratch
• Act as a technical advisor and embedded engineer, translating ambiguous business problems into well‑scoped AI systems with clear success metrics
• Instrument and monitor deployed agents in production — you’re on‑call for what you ship, and you treat reliability as a feature

• 5+ years of production software engineering experience, primarily in Python or Type Script. Go is a plus
• Production LLM application experience with Anthropic or OpenAI SDKs — agents, structured outputs, tool use, RAG, evals, batch processing — shipped, not demoed
• Forward‑deployed instinct: engineering, developer relations, or solutions engineering experience
• Strong evaluation discipline with the ability to define and defend exit criteria using Lang Smith, Braintrust, or equivalent tools
• Experience building multi‑step tool‑using agents with planning, error recovery, and human‑in‑the‑loop design in production environments
• Experience with RAG pipelines, embeddings, hybrid search, and the judgment to determine when retrieval improves outcomes
• Experience building MCP servers, function‑calling schemas, and sandboxed execution environments
• Strong understanding of token budgets, model tier trade‑offs, and AI cost/latency optimization strategies
• Experience integrating REST APIs, Graph
  
  QL, webhooks, OAuth/SAML authentication (especially Okta), and event‑driven architectures
• Cloud‑native engineering experience with GCP or AWS, including Terraform, containers, secrets management, logging, metrics, and alerting
• Strong SQL and data engineering experience with modern warehouses, ETL/ELT pipelines, schema…