Embed with enterprise customers to understand their workflows, data environments, and operational constraints — and build AI solutions that fit their reality, not a sanitized sandbox
Own implementations end-to-end: from scoping and solution design through integration, testing, deployment, and handoff
Rapidly diagnose technical blockers — messy data, broken integrations, edge cases, legacy system quirks — and solve them yourself without waiting on a queue
Design, build, and orchestrate multi-step AI agents that automate complex workflows across enterprise systems
Work with frameworks like LangGraph, LangChain, or similar to architect reliable, production-ready agentic pipelines
Apply sound judgment about what should be automated vs. what requires human-in-the-loop, and design accordingly
Continuously tune agent behavior based on real-world usage patterns you observe in the field
Build custom integrations, connectors, and data pipelines to bridge enterprise tech stacks with AI infrastructure
Work across the stack — APIs, vector databases, LLM APIs, cloud infrastructure, and front-end surfaces — to deliver complete, working systems
Write clean, maintainable code that others can build on top of
Identify patterns across customer engagements that signal genuine product opportunities — and advocate for them with engineering and product teams with precision
Contribute to building internal tooling and repeatable delivery assets (deployment templates, agent blueprints, evaluation frameworks) that make the next implementation faster
Optionally: take ownership of building product features or internal tools that emerge from your field insights
Requirements
3+ years of software engineering experience, with at least 1–2 years focused on AI/ML systems in production environments
Hands-on experience building with LLMs (OpenAI, Anthropic, Gemini, or open-source models) — prompt engineering, RAG pipelines, fine-tuning, evaluation
Experience designing and deploying agentic AI workflows — multi-step reasoning, tool use, memory, planning
Strong programming skills in Python; comfortable with APIs, cloud services (AWS/GCP/Azure), and enterprise databases
Proven ability to work directly with enterprise customers or technical stakeholders — you can translate complexity into clarity
End-to-end ownership mindset: you're not done when the code ships; you're done when the customer succeeds
Strong Signals We'll Look For
You've debugged a production AI system that was misbehaving inside a customer's environment and fixed it under pressure
You've built something agentic that actually ran in production — not just a toy demo
You have opinions about AI system design, and you can back them up with experience
You've identified a customer problem that became a product feature
You've shipped something end-to-end — a product, a tool, a system — that you designed yourself from scratch