Engineering Department (all applications) - m/w/d

Responsibilities

• Build resumable streaming infrastructure where users can disconnect mid response and pick up exactly where they left off.
• Design retry logic that continues generation from the exact stopping point when providers fail mid stream.
• Build and extend our workflow automation system where agents can pause for human input, extract structured outputs from free form conversations, and chain together actions across hundreds of integrations.
• Implement distributed loop orchestration with Redis coordinated concurrent execution and race condition prevention.
• Solve problems like context window management while preserving tool call integrity, intelligent message reduction strategies, and token refresh callbacks that happen transparently during long running executions.
• Own how AI systems manifest in the product end to end.
• Design and implement novel ways for humans and agents to collaborate, review reasoning, steer generation, and compose workflows that do not yet have established UI patterns.
• Design and maintain the infrastructure that runs Langdock across dozens of environments on Azure, AWS, GCP, and bare metal servers.
• Own domains end to end. We ship daily.

Requirements

• We bring all leading AI models into one secure, model-agnostic platform and make them usable across entire organizations.
• WHAT YOU WILL ACTUALLY DO Ship features across our AI engine that orchestrates the newest model releases from all relevant AI labs through a unified abstraction layer.
• Invent new interaction paradigms for AI native software. Go beyond chat.
• WHAT MAKES THIS ROLE DIFFERENT You will work on a production grade AI platform with enterprise features like multi tenant observability, connection pooling for external services, sandboxed code execution with secret redaction, queue based scaling, and middleware that normalizes provider specific API differences.
• You do not just use AI tools daily.
• TypeScript, React, Node, databases, queues.

Experience