• 5+ years of professional software engineering with Python as your primary language, and a track record of building and running production-grade software – including being the person who fixed it when it broke.
• Python craft: strictly typed, well-structured code that goes well beyond scripting or notebooks. You model your data at every boundary – Pydantic, attrs or similar – and you're comfortable working within an established framework. 
• A production mindset: you think about maintainability, testing, telemetry, and debugging, and you’re comfortable handling production issues and contributing to root cause analysis (RCA). 
• Deep, practical async Python: you have built services on asyncio and you know what it buys you and what it doesn't – async generators, cancellation, and why a single blocking call inside a coroutine stalls everything around it. Micro-services and event-driven architectures on top of that.
• A feel for what goes wrong across a network boundary: timeouts, bounded retries with backoff, idempotency. You read an outbound call with no deadline on it as a defect rather than a detail. 
• Proficiency in designing and consuming RESTful APIs. 
• An affinity for AI/ML: curiosity about or hands-on experience with LLMs and agentic systems is a strong plus, though neither is required – we care more that you engineer well than that you already know this stack. Deep ML specialism is not needed. 
• Strong communication skills in English.

Nice to have: 

• Supply chain or logistics domain knowledge – the cherry on top, not a requirement. 
• Willingness to dig into the Java codebase of our core product to understand what functionality is already available, even before it’s exposed via an API. 
• Experience with CI/CD (GitHub Enterprise) and Infrastructure-as-Code (Terraform, Pulumi). 
• Familiarity with cloud AI services (e.g., SageMaker, Bedrock, Azure ML).
• 
  
•