Software Engineer

As a software engineer at Atira, you will work on the systems that make frontier AI usable in production: The platform, integrations, and infrastructure that let agents operate reliably inside complex industrial environments, and you'll touch the AI pipelines and harnesses themselves whenever the problem demands it. The backbone you build handles thousands of concurrent workflows, processes documents with thousands of pages, scales to millions of concurrent LLM requests, and writes back into the SAP, Salesforce, and PLM systems our customers run on.

Misreading or wrongly interpreting a spec can cost a manufacturer six figures, so the challenges are real and the margin for error is thin.

Where you trained matters less than what you've built and how honestly you can describe what broke. Atira is at a stage where you must own problems end to end, because what you ship today is what our customers use tomorrow.

The key addition is the second sentence: "AI is not a separate track here, it's woven through the entire stack ... you'll touch the AI pipelines and harnesses themselves whenever the problem demands it."

That line does three things at once:

• Signals to candidates that this isn't a "backend role at an AI company" where they'll be walled off from the interesting model work.
• Sets up the boundary correctly: SWEs touch AI pipelines when the problem demands it; AIEs own them. That's honest about the overlap without erasing the role distinction.
• Makes the JD attractive to the kind of generalist SWE who wants AI exposure but doesn't want to be a pure AI engineer, which is probably your actual target candidate.

If you want it shorter, you could compress that sentence to: "AI runs through the entire stack here, and you'll work on the pipelines and harnesses themselves whenever the problem demands it." Same idea, one clause lighter.

As a software engineer at Atira, you will work on cutting edge tech such as agent harnesses, runtimes, evals and pipelines balancing deterministic accuracy with agentic agency. You build AI Agents that reliably process documents with thousands of pages and scale our system to handle 1M+ LLM calls per day. Misreading a spec can cost a manufacturer six figures, so the challenges are real and the margin for error is thin. Where you trained matters less than what you've built and how honestly you can describe what broke. Atira is at a stage where you must own problems end to end, because what you ship today is what our customers use tomorrow., * Build full-stack features, infrastructure and AI pipelines: You'll design, implement and optimise across our entire stack: From building new functionality in close collaboration with forward-deployed engineers and customers to scaling and optimising our infrastructure across core and AI systems.

• Own the platform the agents run on. The backend services, APIs, queues, and infrastructure that let agents execute reliably across thousands of concurrent customer workflows.
• Unlock industrial scale. 1000s page inbounds, excel pricing lists, technical drawings, CAD files, ERP tables, configurator rules - industrials live in multi-modal, data heavy environments where reliability is crucial. You will be responsible to continuously scale our systems and infrastructure: 1000s of users, high-throughput ETL pipelines, millions of concurrent LLM requests, tenant-isolated environments with no silent failures, cost blowouts, or latency explosions.
• Own the integrations layer and enterprise infrastructure. Industrials live in SAP, Salesforce, custom configurators, PLM systems, and other databases. You design the connectors, abstractions, and write-back logic that agents can interact with. Multi-tenancy, enterprise deployments and state-of-the-art IT security measurements included.
• Build the developer platform. CI, deployment, dev environments, end2end tests, observability and internal AI tooling. Your work will 10x our development speed and unblock every engineer.

• Real engineering fundamentals. Strong skills in Python, TypeScript, or a comparable language. You can read someone else's production code, understand why it was built that way, and ship changes that don't make it worse.
• System design & infrastructure excellence. You understand how to design and build scalable systems from an architecture and infrastructure perspective.
• You've gone further than most with AI-assisted development. Whether it's Claude Code, Cursor, Codex, or your own setup, you've invested seriously in building the skills, context, and workflows to get maximum leverage from coding agents. This isn't a nice-to-have. We think engineers who have mastered this are fundamentally more productive, and we hire accordingly.
• You are a strong technical decision maker that knows how to scale systems into millions of calls per day, while being able to explain the pros and cons of every solution. Yes, every solution: there is no free lunch.
• You've demonstrated ownership and initiative: a company you started, a serious open-source contribution, a side project that demand drove to production scale, a role where you owned a technical outcome others would have delegated. We want to see evidence that you build things without being told to.
• Degree in Computer Science, Information Systems, AI, or a related technical field from a strong technical university.
• Full working proficiency in English. German is not required

• Real ownership: you'll be one of the early team members and shape both product and architecture as well as work closely with customers.
• Strong peers: work with people who have built and shipped AI systems in industrial environments at scale before, and the chance to learn directly from them.
• Impactful work: your code is deployed all across Europe's and US industrial backbone, affecting how complex products and services are sold and configured worldwide.
• Competitive salary & equity package and additional benefits including Wellpass, JobRad and company dinners.
• Most importantly: Collaborate & thrive in a high-performing but caring culture. We will own the industrial sales function but will do so with modesty and integrity.

What success looks like after a few months

• You've shipped impactful features and meaningful improvements to the platform, integrations, infrastructure or pipelines that are running in production across multiple customer deployments.
• You own a core system (feature, platform/infrastructure component, integration, observability stack) and the team trusts your judgment on how it should evolve.
• When something breaks in production, you can trace issues across services, databases, queues, and third-party systems, and fix them end to end.
• You've built internal tooling or developer-experience improvements that made FDEs and AI engineers measurably faster.
• Your architectural decisions have held up under real load and you can explain the tradeoffs you made and what you'd do differently.