Software Engineer II, AI Platform

Flexport is using software—and now AI—to accelerate the transformation of the freight forwarding industry.

Our AI Platform team in Amsterdam is building the intelligence layer that will make every person at Flexport exponentially more effective. We're not just building chatbots—we're creating autonomous agents that eliminate toil, catch errors before they cascade, and handle the "long tail" of logistics operations that traditional software could never economically address. This isn't hypothetical—we already have agents saving thousands of hours per week. You'll be building the platform that multiplies that impact across the entire company.

This is a rare opportunity to work at the intersection of AI and a massive real-world problem space. You'll build systems that directly impact how goods move across 112 countries, touching everything from customs compliance to document processing to incident response.

What You'll Build:

Automation Engineering Platform

• Build the infrastructure that lets operators create AI-powered workflows without writing code—parsing emails against SOPs, extracting data from documents, creating smart work queues, and deploying ambient quality agents that catch problems before they become incidents
• Create the tooling that transforms subject matter experts into automation engineers
• Create AI powered safeguards and security, that monitor usage and warn (or even intervene) when we have potential problems.

Secure AI Infrastructure

• Build a production-grade, internal AI chat surface with granular MCP (Model Context Protocol) integration, giving different teams secure access to the right internal tools and data
• Create an MCP gateway that safely exposes internal services to client-side AI agents, enabling developers to use tools like Claude Code against Flexport's systems
• Navigate the rapidly evolving auth and security landscape of AI tooling—you'll be solving problems that don't have Stack Overflow answers yet

Agent Templates & Deployment

• Turn our most successful AI agents—security monitoring, data repair, compliance checking, incident response—into reusable templates
• Build the deployment infrastructure that makes it trivial to spin up new specialized agents for different use cases
• Design systems that make agents observable, debuggable, and improvable over time

Why This Role Is Special:

• You're early: AI in enterprise is still being figured out. The patterns you build here could become industry standards
• Real problems: You're not optimizing ad clicks—you're helping businesses move physical goods across borders more efficiently
• Autonomy: You'll own entire projects from conception to deployment

Learning: You'll work alongside people who have built successful AI agents in production and know what works (and what doesn't)

You Should Have:

• ≥ 3 years of software engineering experience, and having built at least one or two production LLM applications
• Full-stack capability: You're comfortable deploying services on Kubernetes/AWS, designing backend systems for LLMs, and building frontend applications when needed
• Deep LLM experience: You understand agent patterns, workflows, MCP, RAG, prompt engineering. You've worked with frameworks like LangChain, LlamaIndex, or built your own orchestration layers.
• Entrepreneurial mindset: You can take a vague problem ("make compliance checking automatic"), design the full solution, and ship it to production without needing your hand held
• A love of simple, well-tested code that is easy to extend and hard to break—especially important when working with non-deterministic AI systems
• Excellent communication skills: You'll be working directly with operators, product teams, and leadership to understand needs and explain technical tradeoffs
• A bias for action: When you see an opportunity to eliminate toil or catch errors, you build it
• Non-optional: an aptitude for learning quickly in a space where best practices are being written in real-time

Nice to Have:

• Experience with security, especially around API gateways and auth systems
• Experience with workflow orchestration tools (Temporal, Airflow, etc.)
• Familiarity with the logistics or supply chain domain