About the roleFlow is seeking staff software engineers to build AI-powered capabilities that help teams author, review, and manage requirements more effectively. You will work on agentic systems engineer and agentic domain engineer workflows-bringing AI into the center of how teams reason about their systems.
This role sits at the intersection of AI, product, and full-stack engineering: you will take ideas from prototype all the way to stable, observable features in production.
What you'll do- Design and ship AI-powered features such as assisted requirement drafting, consistency checks, impact analysis, and intelligent suggestions for systems and domain engineers.
- Build agentic workflows that help "agentic systems engineers" and "agentic domain engineers" explore designs, simulate changes, and validate requirements.
- Evaluate and integrate language models and related tooling, optimizing for reliability, latency, cost, and debuggability in production.
- Build and maintain the surrounding infrastructure: data pipelines, evaluation harnesses, prompt and model management, observability, and safety/guardrails.
- Work across the stack-from backend integrations and APIs to simple UI hooks-to deliver complete AI features, not just model endpoints.
- Partner with product and customers to identify high-value workflows, run experiments, and iterate quickly based on usage.
About you- 8+ years developing production software, including designing, testing, and operating services at scale in a cloud environment.
- Hands-on experience with modern LLM providers and tooling (e.g., OpenAI, Anthropic, Hugging Face, vector stores, RAG patterns).
- Familiarity with prompt design, retrieval-augmented systems, evaluation methods, and safety/guardrail approaches.
- Ability to reason about tradeoffs between different models, architectures, and deployment patterns and make pragmatic decisions.
- Comfortable working in a high-ownership, fast-paced environment where experiments and iteration are the norm.
Our stack (AI-leaning)- TypeScript/Node.js and Python for AI and backend services.
- Modern LLM APIs and orchestration libraries for building agentic workflows.
- Postgres and other managed cloud services for data and state.
How we work & values- Speed over everything: prototype AI workflows quickly, then harden what works.
- Own, downscope, ship, iterate: one clear owner per feature, from prototype to production.
- Fundamentals done well: evaluation, observability, and safety are part of the first version, not an afterthought.
- Competitive salary and meaningful equity.
- Health, dental, and vision coverage.
- Flexible time off and support for experimentation, learning, and staying current with the AI ecosystem.
