AI Engineer

This is a fully remote position, open to applicants in United States.

📋 Description

• Take ownership of AI products in production, ensuring their stability and observability, with regressions identified through evaluations prior to learners encountering them.

• Transition prototypes from functional demonstrations to fully maintained products, incorporating evaluations from the outset.

• Design the context and tooling architecture for agents, defining what constitutes context, when it applies, and the rationale behind it.

• Develop and sustain service interfaces that agents utilize, including providing application access to the learner’s artifact window.

• Ensure the reliability of long-running, multi-step agents under genuine learner behavior, with well-defined failure modes and recovery protocols.

• Conduct tests on deployed systems using evaluation loops, regression checks, and rapid iterations.

• Collaborate with Anthropic Engineers, AI engineers, product managers, and data scientists to craft effective learning experiences.

⛳️ Requirements

• Significant, production-level experience with Python and FastAPI involving real user interactions.

• At least one deployed AI/LLM application that the candidate can discuss comprehensively, including any failures, diagnosis, and subsequent adjustments.

• Proficient understanding of agent frameworks; experience with LangGraph is preferred, but other equivalent frameworks are acceptable.

• A genuine perspective on context engineering.

• A consistent practice of testing developed solutions.

• AI-native programming skills with proven experience in harness engineering.

• Ability to function effectively within small physical teams as well as larger virtual collaborations.

• A self-motivated approach to continuous communication using web collaboration tools and video conferencing.

• Desirable but not mandatory: Experience with React and a full-stack skillset.

• Desirable but not mandatory: Familiarity with Langfuse or similar observability and knowledge graph tools.

• Desirable but not mandatory: Background in the Educational/Edtech domain.

• Preferred time zone: GMT+2.

🏝️ Benefits

• Ongoing development of skills and capabilities.

• Encouragement to pursue certifications that reflect personal growth.

• Opportunities for virtual work and interaction with colleagues and clients across various international locations.