HybridWork from the Böblingen office two days a week; travel to client sites across DACH.
Software EngineeringAI Application Engineering
About the job
Requirements
Have at least 4 years of software engineering experience, including taking at least one AI or large language model system to production.
Have hands-on experience with retrieval-augmented generation, tool use or Model Context Protocol, and systematic evaluation of large language model output.
Have strong Python or TypeScript skills, plus experience with REST APIs, webhooks, and integrating business systems such as CRM, ERP, or data platforms.
Have experience working directly with client or business stakeholders, from understanding the problem to defining scope.
Have German proficiency at C1 level or higher and fluent English.
Be willing to work from the Böblingen office two days a week and travel to client sites across DACH.
Responsibilities
Own technical delivery of agentic AI use cases from the first discovery workshop through a stable production system.
Build agents and integrations on HubSpot, ERP, and data platforms, with evaluations, monitoring, and clear acceptance criteria.
Debug in real client environments, stay until the solution works, and hand over cleanly to the Managed Services team.
Turn recurring work across clients into reusable connectors, Model Context Protocol servers, agent templates, and evaluation tooling.
Ensure every deployment meets GDPR and EU AI Act requirements before go-live.
Explain trade-offs to client stakeholders and document in English.
Support consultants and the sales team with technical scoping and rapid proofs of concept.
Desired Qualifications
Have experience with HubSpot APIs, SAP, Salesforce, or Microsoft Dynamics integrations.
Have cloud and serverless experience with AWS, Azure, or Google Cloud Platform, and experience with large language model tracing and cost monitoring.
Have a consulting or agency background working with enterprise clients.
You'll be the engineer on the ground at our enterprise clients. You sit with their teams, find the workflows where AI agents create real value, and build them until they run in production, connected to HubSpot and the systems around it. What you learn in the field, you turn into building blocks the whole team ships with.
What You'll Do
Own the technical delivery of agentic AI use cases from the first discovery workshop to a stable production system
Build agents and integrations on top of HubSpot, ERP and data platforms, with evals, monitoring and clear acceptance criteria
Debug in real client environments and stay until it works, then hand over cleanly to our Managed Services team
Codify what repeats across clients into reusable connectors, MCP servers, agent templates and eval tooling
Ship compliant: every deployment you own meets GDPR and EU AI Act requirements before go-live
Translate between business and tech, explaining trade-offs to client stakeholders in German and documenting in English
Support our consultants and sales team with technical scoping and fast proofs of concept
What You Bring
Must-haves
4+ years of software engineering, including at least one AI or LLM system you personally took to production (not only a prototype)
Hands-on experience with RAG, tool use or MCP, and with evaluating LLM output systematically
Strong Python or TypeScript, plus experience with REST APIs, webhooks and integrating business systems (CRM, ERP or data platforms)
Experience working directly with client or business stakeholders, from understanding the problem to defining scope
German at C1 level or higher and fluent English
Willingness to work from our Böblingen office two days a week and travel to client sites across DACH
Nice-to-haves
Experience with HubSpot APIs, SAP, Salesforce or MS Dynamics integrations
Cloud and serverless experience (AWS, Azure or GCP), LLM tracing and cost monitoring
A consulting or agency background with enterprise clients
Your Team
You'll report to the Team Lead AI Engineering Germany, joining the AI Engineering Germany team within Thorit's AI & Engineering Practice. You'll work closely with our AI Engineering team in Lisbon, our UX & UI designers, and the consultants and project managers staffed on your client engagements.
Tech Stack
Python, TypeScript, LLM APIs (Claude and others), agent frameworks and MCP, HubSpot APIs, cloud and serverless platforms, Git and CI/CD, Claude Code, Jira, Confluence, Forecast.
What We Offer
31 vacation days per year (30 plus one extra day for Christmas or a holiday of your choice), well above the statutory minimum of 20
Flex work model: two days in our Böblingen office, the rest where you work best
Learning budget every year for conferences, certifications and courses
MacBook and a full desk setup
Company pension (Allianz) with an employer contribution from your second year, rising with tenure
Business Bike leasing with tax advantage, unlimited private use and full insurance
Lunch subsidy of €5 tax-free per working day (up to €100 per month)
Urban Sports Club memberships
Corporate Benefits discounts from several hundred brands
About Thorit
Thorit is the AI & CRM consultancy for the agent-native enterprise. As a Diamond HubSpot Partner we work with enterprise clients in DACH and the Nordics, building the agent layer on top of HubSpot and the systems around it, then operating it with them. Agentic means software that acts on goals instead of waiting for clicks. For our clients, that means less manual work and more margin.
Ready to Apply?
Send your application via Ashby. A cover letter is optional, but a short description of one AI or integration system you took from a client conversation to production is required.