cargo.one is a Berlin-based AI-native operating system that unifies air and ocean freight rate data to automate sales and procurement for freight forwarders and carriers.
🏆 What's the opportunity? In 2026 we launched the industry's first AI-native operating system for freight: one workspace where freight teams and AI agents work side by side across rate management, procurement, quoting, booking and support. The AI here isn't a roadmap slide — it ships to production and quotes real freight every day. You'll build that system, and you'll own problems end to end: shaping the architecture, writing the spec, shipping the code and watching real freight forwarders use it the same week. In a three-engineer squad, your design decisions are the team's design decisions. And you'll work the way the product works: spec-first, with Claude Code in the loop daily — the interesting part of your job is the thinking, not the typing. Logistics powers the global economy, but the work behind it is still overwhelmingly manual. We're fixing that for 30,000+ users across 172 countries, with customers like Lufthansa Cargo and Kuehne+Nagel, backed by Index and Bessemer. 🌍 You can work remotely from across EMEA (within our approved hiring countries); we meet in person regularly, and can support visa processes in Germany only. The culture is positive, diverse, hard-working and feedback-heavy, with a strong dose of playfulness. 🎬 A snapshot of what you'll be doing - Own problem spaces end to end: you write the spec and the acceptance criteria, agents do much of the typing, and you own the review, the architecture and what ships. - Build AI into the product itself — for example: when an airline quotes a price over email, our AI agents turn that free-text reply into a quote the customer can book in the app. - Make AI trustworthy where a wrong number books real freight: structured outputs, evals, confidence-gated human review, monitoring. - Raise the squad's AI leverage: better specs, better harnesses, better review — your workflow improvements ship to every engineer. - Build self-service tools (Vue.js front-ends on a Python backend) so colleagues help themselves instead of filing requests — often agent-assisted, always yours to design. - Hold the quality bar in a codebase where a lot of the code is generated: automated tests, metrics, tracing, alerts, and real review discipline. - Support your users hands-on, challenge assumptions, and ask "why" before you build. 🛠 What you'll need - 5+ years of backend experience building web applications, with strong architecture instincts (request/response flow, service integration). - Strong SQL, including data consistency, indexing and race conditions. - A spec-driven way of working: you plan and define acceptance criteria before you code, and write clearly (we're remote-first). - You're fluent in the modern AI engineering stack — LLMs, agents, MCP, coding assistants, agent harnesses — and you use it daily. You ship with Claude Code or an equivalent, and you've built or wired AI into a real system. - More importantly: judgement. You validate, critique and refine what models produce, you know where they go wrong, and you know when not to reach for AI. - Product sense and empathy for the people using what you build. - Low-ego collaboration: you give and take feedback and work self-directed. Bonus points: you've built your own agentic tooling, rather than just consuming it; our stack (Python 3, Flask, gRPC, PostgreSQL, Redis, Celery, Kubernetes on GCP), Vue.js, or experience raising a team's AI fluency. 🤖 The AI here is real - An LLM pipeline in production turning airlines' free-text email replies into structured, bookable quotes. - AI review on every merge request in CI, and agent-built internal tools on MCP. - Every engineer ships with Claude Code — real autonomy, no innovation theater. - Even the interview reflects it: you'll drive your own agent live, and we assess how you work with it — not whether you can type. ❤️ We think you'll love working here if - You take ownership and get a kick out of exceeding expectations. - You'd rather opt for the ambitious option to grow than take the easy one. - You see giving and receiving feedback as core to how you work. - You voice opinions and back them with data. - You experiment with new ideas and hunt for process improvements. ✨ What we offer - €650 home office budget, plus your choice of MacBook or Dell laptop delivered before day one. - Co-working access through our partner Desana. - 24 paid vacation days on top of your country's public holidays, plus special leaves. - Offsite fun and team trips: we meet twice a year as a company. - Permanent full-time employment through our partner Remote.com http://Remote.com. 🤔 Why cargo.one http://cargo.one? Since 2018 we've been building the data foundation that makes AI actually work in freight — supply, rates and booking connectivity nobody else has. That foundation is why our agents do real work instead of demos, and why engineers here solve problems the rest of the industry is still writing thought-leadership posts about. We hire for talent. If you don't tick every box but think you'd be a great match, reach out anyway; we'd love to hear from you.