Founding AI Engineer Onsite (Munich, Germany)

S27a; München; Vor Ort; EUR 95.000 - 150.000; Vollzeit

Zusammenfassung

S27a is seeking a hands-on software engineer to define and build the AI Worker platform that operates in live logistics environments. You will own architecture, context management, and observability to ensure reliable production-grade systems across multiple integrations and tools.

You will ship, measure, and refine workflows across email, telephony, ERP, and APIs, while shaping engineering standards and product direction with the founders. Munich office—onsite, fast-moving, impact-driven.

Qualifikationen

• Strong software engineering fundamentals and production-grade systems.
• Hands-on with LLM systems in production: workflows, context management, evaluations, observability.
• Experience with distributed, asynchronous systems: queues, retries, idempotency, concurrency.

Aufgaben

• Own the architecture and evolution of the core AI Worker platform across backend services, agent workflows, integrations, data, and infrastructure.
• Build reliable, long-running AI workflows that operate across inboxes, phone, TMS, ERP, EDI, SFTP, APIs, and event queues.
• Develop production-grade agent systems: evaluations, tracing, feedback loops, guardrails, retries, fallbacks, and escalation.
• Tackle challenges of applied AI: ambiguous inputs, changing context, tool use, state management, latency, cost, and failure recovery.
• Turn customer deployment patterns into reusable product capabilities rather than one-off fixes.
• Make deliberate tradeoffs between speed and durability; ship simple versions first, then plan rebuilds as needed.
• Shape tech direction with founders; defend decisions with data and production-grade thinking.
• Help set engineering standards, from design to AI-tool usage for faster delivery.

Kenntnisse

Python

JavaScript/TypeScript

SQL

REST APIs

Data modeling

System design

Tools

Docker

CI/CD

Cloud infrastructure

Monitoring

Incident debugging

Jobbeschreibung

The Role

Work directly with the founding team to define and build the technical foundation behind our AI Workers. These systems operate inside live logistics workflows across email, telephony, and enterprise software, where reliability matters and edge cases are the product.

You will own the hard parts underneath the features: agent architecture, context and state, evaluations, observability, execution reliability, data flows, and the platform other engineers build on.

This is a hands-on role. You will write production code every day, but your job is bigger than shipping a backlog. You will decide which abstractions survive, where we need rigor, what we build next, and how the engineering team operates as we grow.

This is not a research-only role, and it is not an architecture role where you stop coding. You build, ship, measure, fix, and keep raising the bar.

What You’ll Do

• Own the architecture and evolution of the core AI Worker platform across backend services, agent workflows, integrations, data, and infrastructure.

• Build reliable, long-running AI workflows that make decisions and take action across inboxes, phone calls, TMS, ERP, EDI, SFTP, APIs, and event queues.

• Develop the systems that make agents dependable in production: evaluations, tracing, feedback loops, guardrails, retries, fallbacks, and human escalation.

• Work through the difficult parts of applied AI: ambiguous inputs, changing context, tool use, state management, latency, cost, and failure recovery.

• Turn patterns from customer deployments into reusable product capabilities instead of one-off fixes.

• Make deliberate technical tradeoffs between speed and durability. Ship the simple version first, then know when the system needs to be rebuilt properly.

• Shape product and technical direction with the founders. Challenge assumptions, propose better approaches, and own decisions through production.

• Help set the engineering standard for the company, from system design and code quality to how we use AI tools to build faster.

• Help hire and develop the engineering team as we grow.

Who You Are

• You take ownership of unclear, important problems. You do not wait for a ticket, a complete spec, or someone else to define the path.

• You have strong technical judgment. You can simplify a system without being careless, and add structure without slowing everyone down.

• You think like a product builder, not only an engineer. You care whether the system solves the real operational problem, not whether the implementation looks clever.

• You move fast and iterate constantly, but you do not hide behind prototypes when customers depend on the system.

• You build with AI, not just around it. AI-native development is how you think and work.

• You are low ego and direct. You can disagree clearly, change your mind quickly, and make the best argument win.

• You are comfortable being close to customers, production incidents, and the messy details of logistics operations.

• You want to be in the room. Munich office, in person. This is not a remote role or a 9-to-5.

What You Bring

• Strong software engineering fundamentals: Python, JavaScript/TypeScript, SQL, REST APIs, data modeling, and system design.

• Experience owning production systems across architecture, implementation, deployment, and operations.

• Hands-on experience with LLM systems in production: agent workflows, tool use, prompt engineering, structured outputs, context management, evaluations, and observability.

• Strong understanding of distributed and asynchronous systems, including queues, retries, idempotency, concurrency, background jobs, and failure handling.

• Experience with Docker, CI/CD, cloud infrastructure, monitoring, and production incident debugging.

• A track record of shipping products or systems that real users depend on.

• Bonus: experience with enterprise integrations, voice systems, email automation, logistics, supply chain, or freight.

Process

We move fast. Expect a decision within one week.