Teams at Onit

Supports legal AI product development through legal workflow research, risk assessments, compliance documentation, prompt engineering, jurisdictional regulatory research, and independent validation. The role evaluates AI outputs for legal, ethical, privacy, trust, and safety requirements; maintains risk registries and audit logs; prepares model factsheets and privacy statements; and collaborates with product and engineering teams to deliver trustworthy, compliant AI solutions.

Build and ship production agentic AI features using LLM APIs, tool-calling, multi-agent workflows, RAG, and context engineering. Establish evaluation frameworks, regression gates, monitoring, observability, and safety guardrails. Develop scalable backend services and integrate AI into cloud and event-driven architectures. Partner with product and domain teams to prototype solutions, document failure modes, and improve quality, cost, latency, reliability, and security through iterative experimentation.

Develop and execute UI, API, load, and performance tests using Playwright or Selenium with Artillery. Write automated tests in JavaScript or Java, validate data with SQL, integrate testing into CI/CD pipelines, analyze failures, perform root cause analysis, and collaborate with development and product teams. The role also involves testing AI-driven features and using AI coding tools to accelerate test development and debugging.