Software Engineer II: Applied AI

The $8.6 trillion freight forwarding industry runs on phone calls, faxes, and emails. At Flexport, we're changing that with software—and now, with AI, we're accelerating the transformation by orders of magnitude.

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 $19 billion of 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 core software engineering experience, with a proven track record of shipping production-grade LLM applications rather than just notebooks or research projects.
• Full-Stack Capability: You are an engineer who can own the entire lifecycle of a feature. This means you are comfortable deploying services on Kubernetes/AWS, designing robust backend systems for LLMs, and building the React-based frontend applications that operators actually use.
• You understand the engineering patterns required to make AI work at scale. You have deep experience with agent patterns, orchestration layers, RAG, and the Model Context Protocol (MCP).
• You have worked with tools like LangChain or LlamaIndex, or more importantly, you have built your own custom orchestration layers to solve specific production needs.
• You have a passion for simple, well-tested code that is easy to extend and hard to break—this is essential when building reliable systems around non-deterministic AI.
• You can take a vague problem, such as "automate compliance checking," and independently design the full system architecture, write the code, and ship it to production.
• You have a demonstrated ability to master new technologies quickly, which is critical in the AI space where best practices and protocols 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