AI engineer

We are seeking a Senior AI Engineer to join an open-source engineering governance initiative for AI coding agents, developed within the R&D division of a global digital engineering and technology services company. In this role, you will shape how AI coding agents such as Claude Code, Cursor, Copilot and Codex work inside real engineering teams, designing the skills, structured context and retrieval mechanisms that load a team’s shared instructions, architecture and constraints into every agent session, so the agent follows the team’s engineering discipline instead of guessing from a few open files.

• Agent Skills & Context Engineering: Design and build skills for AI agents, along with structured context that delivers the right rules, knowledge, workflows and constraints to MCP-compatible coding agents at the right moment.
• Meta-Prompting & Model Behavior: Develop formalized, structured prompting approaches that teach agents how to think about engineering problems rather than prescribing step-by-step actions.
• Retrieval & Platform Engineering: Build RAG pipelines, semantic retrieval and context construction on top of production-grade APIs and microservices in Python and TypeScript.
• Evaluation & Observability: Create automated evaluation and testing systems for LLM-based features and use LLM observability to measure and improve agent behavior across open-source and commercial models.
• Product Ownership & Technical Leadership: Take an active part in product strategy, decision-making and problem-solving with end-to-end ownership of outcomes, while mentoring engineers and guiding technical direction.

• Commercial experience in AI/LLM engineering, including hands-on practice building skills for AI agents and a strong understanding of how LLMs work.
• Meta-Prompting & Context Engineering: Hands-on experience controlling model behavior with structured prompting and designing context mechanisms for rules, knowledge, skills, workflows and constraints.
• LLM Engineering: Strong experience integrating and configuring models through APIs/SDKs such as OpenAI or Anthropic, working with open-source LLMs, and using LLM orchestration frameworks and LLM observability tooling.
• RAG & Semantic Retrieval: Practical experience implementing retrieval pipelines and context construction with vector databases.
• Python & TypeScript Engineering: Strong development skills in Python and TypeScript/JavaScript, with production-grade API and microservices design, Docker and CI/CD.
• AI Quality Engineering: Experience building automated evaluation and testing systems for AI/LLM applications.
• Leadership & Collaboration: Technical leadership and mentoring experience, confident decision-making and effective teamwork.
• Будет плюсом: Multi-Agent Systems: Experience with multi-agent systems and agent orchestration, as well as familiarity with MCP (Model Context Protocol) or similar standards; Advanced Evaluation: Knowledge of LLM-as-Judge methodologies and A/B testing frameworks for AI systems; Search & Embeddings: Deep knowledge of semantic search and embedding models; Developer Tooling: Experience building developer tools, SDKs or IDE integrations, and understanding of progressive disclosure patterns in AI applications; Security Integration: OAuth/OIDC integration experience

• Условий в вакансии нет