Full Stack, AI-Native Engineer

What You’ll Do

• Use AI agents as integral tools for exploration, debugging, code review, and code generation across a large polyglot monorepo, and help build the agent skills, subagents, and tooling that make the whole team faster.

• Build features end-to-end across the Impart platform: Go API services, PostgreSQL and ClickHouse data layers, OpenAPI-first REST APIs, and our React/TypeScript web console.

• Design backend systems that process and analyze high-volume security data, with attention to data modeling, query performance, concurrency, reliability, and operational visibility.

• Build AI-powered product capabilities on Claude, including agentic tool-use loops, AI-assisted rule generation, continuous false-positive analysis, and detection explanations, with guardrails that keep humans in control of enforcement.

• Extend the Impart MCP server so customers’ AI agents can investigate threats, tune rules, and set up alerting directly through natural language.

• Test - build evaluation pipelines and tools that measure AI outputs against ground truth for accuracy, latency, and cost, so LLM features meet functional and performance standards before and after they ship.

• Ship features that secure AI itself: LLM and MCP traffic protection, prompt-injection detection, and AI asset discovery across customer environments.

• Deploy and operate what you build in cloud-native environments (Docker, Kubernetes, AWS, Pulumi), with the metrics and observability to know it’s working.

• Drive projects end-to-end: spec, build, test, deploy, measure, and iterate.

• Contribute to our engineering culture of curiosity, rigor, and technical excellence.

• In person in San Francisco preferred

What You’ll Bring

• You have 2–5+ years of software engineering experience, or you’re a strong recent graduate, with a track record of shipping full stack features from database to UI.

• Strong proficiency in Go, with solid experience in API design, relational data modeling, SQL, concurrency, performance, and production debugging. Comfortable building product interfaces in TypeScript/React.

• Hands-on experience building with LLMs: prompt design, tool use/function calling, structured outputs, embeddings, and evals.

• An AI-native workflow: you already use agentic coding tools daily to move fast, balancing experimentation with sound engineering judgment on when to trust the tools and when to dig in yourself.

• Experience deploying production software in cloud-native environments (Docker, Kubernetes, modern CI/CD, Terraform/Pulumi).

• Excellent collaboration and communication skills; comfortable in a fast-moving, low-ego engineering culture.

Nice to Have

• Experience with Claude Code, including authoring skills, subagents, hooks, or MCP servers.

• Experience with the Model Context Protocol (MCP) or AWS Bedrock.

• Experience with ClickHouse or other columnar analytics databases.

• Background in API/application security, bot protection, or observability tooling.

• Familiarity with OpenAPI-first development and code generation.

Why You’ll Love Working Here

• Build agentic AI into a runtime security platform, and build the security that protects AI applications and MCP servers.

• Work on cutting-edge technology at the intersection of security, observability, and AI.

• Collaborate with an experienced, thoughtful team that values deep technical understanding and craftsmanship.

• A collaborative, in-person environment in the heart of San Francisco’s South Park / SOMA district.

• The opportunity to shape how modern developers build, ship, and secure software in the AI-accelerated era.