Full Stack Engineer, Applied AI

This confidential client is hiring a Full Stack Engineer, Applied AI to help build the intelligence layer of a platform that turns customer intent into real-world operational execution. This is an engineering-focused role that productizes frontier models into reliable production systems — agent loops, retrieval pipelines, internal tools, evals, observability, and end-to-end product workflows. This is not an ML research or model-training role; you will build AI systems that reason over operational context, dispatch work to human operators, detect and mitigate risk, and take action in the real world.

• Build full-stack AI product features end-to-end across React, TypeScript, backend services, database schema, agent logic, and evals.
• Create agents that receive customer requests and dispatch operational work to appropriate human operators.
• Design and implement agent loops that plan multi-step actions, call internal tools, request help when needed, and recover when reality changes.
• Build retrieval and structured-context systems (RAG) that ground agents in operational data.
• Surface operational risks proactively and create automated or assisted remediation before issues impact customers.
• Instrument evals, monitoring, and production observability to measure agent quality and catch regressions.
• Iterate on prompts, tool definitions, model choices, and agent architecture using production telemetry.

Searching for Development & Programming roles that provide visa sponsorship? Connect with international employers through Development & Programming Jobs with Visa Sponsorship opportunities actively seeking talented professionals.

• Partner closely with design and operations to decide what to automate, assist, or leave human-in-the-loop.
• Help define shared AI engineering patterns and best practices that future products can leverage.

• 2+ years of full-stack engineering experience with strong product instincts and ownership from UI to backend to schema.
• Strong TypeScript experience across frontend and backend; experience with React and Node is expected.
• Practical experience shipping AI-driven product features in production (LLM APIs, prompting, retrieval, tool use, structured outputs).
• Working knowledge of retrieval-augmented generation (RAG), retrieval pipelines, agent patterns, and evaluation workflows.

Explore our comprehensive directory of visa sponsorship jobs from employers worldwide who are ready to sponsor talented international professionals.

• Experience with Postgres and Redis or similar databases/caches; comfortable designing schema and backend services.
• Strong judgment about what should be automated vs. assisted vs. human-in-the-loop.
• Comfortable handling messy, real-world workflows: noisy inputs, incomplete data, and systems coordinating human operators.
• Bias toward shipping, measuring in production, learning fast, and iterating quickly.
• Comfortable in an early-stage environment with ambiguity, autonomy, and limited process.
• Interest or experience with tools and platforms such as OpenAI, Anthropic, Claude, Cursor, Azure, custom agent loops, and AI-assisted development workflows.
• Bachelor’s degree or equivalent practical experience preferred.

• Salary: $200,000 - $250,000 per year (base range; final offer may vary based on experience, location, and applicable state/local requirements).
• Employment type: Full-time.
• Workplace: Hybrid (San Francisco Bay Area).
• Visa sponsorship: This role offers visa sponsorship.
• This position is posted on behalf of a confidential client (the hiring organization will make final employment decisions).
  

CareerTakes and our client are Equal Opportunity Employers committed to building a diverse and inclusive workforce. We prohibit discrimination or harassment of any kind. To support a fair and efficient hiring process, AI tools may be used to assist with application review or resume screening. These tools do not replace human decision-making. Final hiring decisions are made by people.

If you have questions about how your data is used, please contact us directly.