Senior AI Platform Engineer

- Design and implement the orchestration layer that turns natural-language intent into reliable execution.

- Build systems for entity resolution, context assembly, connector orchestration, and evidence retrieval.

- Develop reusable Skills that encapsulate business workflows and domain expertise.

- Build routing systems that intelligently coordinate connectors, tools, deterministic logic, and multiple language models.

- Own features end-to-end, from runtime capability to user-facing experience.

- Develop evidence-backed reasoning with citations and traceability that users can actually see and verify.

- Build evaluation frameworks that continuously improve quality.

- Implement permission models, freshness validation, and action policies that work transparently for end users.

- Build prototypes, validate ideas, and rapidly iterate with customers.

- Write high-leverage code that enables entire product areas.

- Collaborate closely with Product, Design, Sales Engineering, and Customer Success to turn ambitious ideas into production systems.

- Experiment with new agent architectures while maintaining production-grade reliability.

- 7+ years of software engineering experience building production systems.

- Deep understanding of distributed systems and backend architecture.

- Experience building AI applications using LLMs, agents, RAG, MCP, or modern AI frameworks.

- Strong software engineering fundamentals including APIs, concurrency, testing, and system design.

- Experience building developer platforms, orchestration systems, or workflow engines.

- Ability to rapidly prototype while maintaining production quality.

- Strong product instincts and comfort operating in ambiguous environments.

- Experience building AI agents or agent frameworks, including evaluation systems.

- Familiarity with orchestration frameworks (LangGraph, Temporal, MCP, or similar).

- Experience with retrieval systems, vector search, or knowledge graphs.

- Experience building developer tools or platform infrastructure.

- Experience with data infrastructure including Kafka, Iceberg, Postgres, Spark, or modern data warehouses.

Success isn't measured by the sophistication of the prompts. It's measured by whether the runtime becomes more trustworthy every week. You'll build systems that:

- Automatically assemble the right context and choose the correct connectors and Skills.

As our runtime evolves, users should think less about tools and more about outcomes. The runtime should handle the rest.

Every company will soon have AI agents. Most will struggle with trust. Agents need more than models: they need context, evidence, permissions, orchestration, and reliable execution.

Airbyte is building that runtime. If we succeed, every AI agent built on Airbyte will inherit those capabilities automatically, allowing developers to focus on solving business problems instead of rebuilding infrastructure.