About Complete Robot
We are an early-stage, venture-backed company currently operating out of San Francisco, building AI-enabled systems for automated physical environments.
The company works at the intersection of AI agents, connected devices, user experience, machine data, and real-world commercial operations. Complete Robot Inc. was founded by Tomasz Stańczak, founder of Nethermind and former Co-Executive Director of the Ethereum Foundation.
To learn more about the company, vision, and the technology we're building, visit the Complete Robot website.
About this role:
We are looking for an Agent / AI Engineer to build AI workflows that interact with real-world systems, data, users, and operational tools.
This is a production AI role, not a generic chatbot role. The systems must be useful, reliable, measurable, safe, and connected to real products and real environments.
For candidates who can start immediately, there may be an opportunity to join the founding team in San Francisco, stay with the team, and build together from a prestigious San Francisco accelerator environment. This is subject to availability, travel and visa requirements, role fit, and mutual agreement.
Responsibilities:
Build AI agents and workflows for customer-facing and operator-facing use cases.
Design tool-use patterns that allow AI systems to interact safely with backend systems and device data.
Build retrieval, structured outputs, evaluations, prompts, guardrails, monitoring, and fallback behavior.
Work with product and design to create trustworthy AI user experiences.
Work with backend and edge engineers to connect AI workflows to real system data.
Measure AI performance in terms of task completion, accuracy, safety, user experience, and commercial value.
Support demos, pilots, and iteration based on customer feedback.
Requirements:
Experience building AI, LLM, agentic, conversational, recommendation, or automation systems.
Strong programming ability in Python, TypeScript, or similar.
Experience with LLM APIs, tool calling, retrieval, structured outputs, evaluations, and production AI workflows.
Understanding of prompt engineering, RAG, agent orchestration, AI evaluation, and common AI failure modes.
Ability to build systems that are reliable, useful, and maintainable.
English fluency required.
Must Haves:
Experience with AI agents for commerce, customer support, retail, payments, kiosks, robotics, voice interfaces, or physical-world systems.
Experience with personalization, recommendations, or conversational UX.
Experience with permissioning, user authentication, transaction approval, or safety-critical workflows.
Experience with multilingual AI systems.
Experience with speech, audio, multimodal AI, or on-device AI.
Success in the first 90 days:
You build the first working AI workflow for a product demo.
You connect AI workflows to relevant data and safe system actions.
You create an evaluation framework for usefulness, accuracy, and safety.
You help produce compelling demos for partners and investors.
Success in the first 6-12 months:
AI workflows are running in real pilot environments.
The AI layer has guardrails, evaluation, monitoring, and measurable value.
You become a key owner of the company's AI systems.