Senior AI Engineer - LLM Agent Development

Build and maintain the AI agent layer on top of Sellforte's marketing analytics and measurement — LLM agents customers can ask, trust, and act on.

Hi — and thanks for considering Sellforte for your next move. 👋

We help retailers and consumer brands grow by making smarter, data-driven marketing decisions. Sellforte has been a data science company from the start: our product runs on real modelling, not vanity metrics. Since 2017 we have built a Marketing Mix Modelling (MMM) solution that marketing teams and agencies use to measure and improve what their marketing and promotions actually do.

Now we are entering a new phase. Our customer base is growing internationally, and we are extending the product beyond traditional MMM toward more granular optimisation, broader decision support, and more automated marketing workflows. A big part of that is a new AI and agent layer on top of our analytics, so customers can ask questions, act on the answers, and trust them — all in plain language, on their own terms.

Our product is a data, modelling, and analytics pipeline that runs under every customer engagement. Now we are building the layer on top of it: LLM-powered agents that get customers and our own teams to answers and recommendations faster. Our roadmap follows where software is heading — increasingly headless and API-first, with capabilities exposed so any interface or agent can consume them — and what we build reflects that. You can see where the AI application interface stands today at demo.sellforte.com.

This is new and fast-moving, and we are looking for an AI Engineer to help design and build it. You will shape how Sellforte's analytics reach people through AI, and you will have a lot of room to shape how the team works as the initiative grows. You will work alongside Emil Kauppi-Hoyer, our Lead AI Engineer leading this initiative, together with our Data Scientists, Customer Success and the product and engineering people building the wider Sellforte platform.

Your main job is building and maintaining the agent layer that makes Sellforte's analytics usable, reliable, and actionable for customers and internal teams. Day to day, that means you will:

• Build and maintain LLM agents, mostly for analytics use cases, both for the platform and increasingly internal one's
• Design agent architectures that handle context efficiently and weigh deterministic tooling against the flexibility of LLMs
• Build and run evaluations to measure agent performance and reliability, and keep improving both
• Develop agent runtimes and tooling, mainly in Python, with some JavaScript and UI backend work where tooling and APIs meet the product
• Integrate agents with internal services, APIs, and — where it makes sense — MCP servers and runtimes
• Contribute to usage tracking and help prioritise work around real customer needs
• Take solutions from prototype to production: deploy, monitor, and maintain agent systems, with an eye on real impact rather than experiments for their own sake
• Use AI-native dev tools (Claude Code, Codex, and the like) to move fast and keep finding better ways to build
• Work closely with Data Scientists, Product, and Customer Success to turn analytics into AI experiences people actually trust

Looking to advance your Development & Programming career with relocation support? Explore Development & Programming Jobs with Relocation Packages that include comprehensive packages to help you move and settle in your new role.

We are looking for someone who pairs strong engineering with good judgment about how LLMs behave. You do not need a research background in ML, but you should be comfortable reasoning about where models are reliable, where they are not, and how to design around that. We are open to both exceptional mid-level and senior candidates — what counts is strong interest and some experience, not years on a CV.

• A solid grasp of LLMs and how they work under the hood
• Hands-on experience building and maintaining LLM agents with agent frameworks, ideally for analytics — comfortable with agent structures, context handling, and the deterministic-vs-LLM trade-offs
• Experience building evals to measure and improve agents
• Strong Python skills and solid software engineering fundamentals, with production systems under your belt
• Hands-on time with agentic coding tools like Claude Code or Codex
• Exceptionally proactive — you are at home at startup speed with limited oversight, comfortable creating direction where there isn't any, untangling open questions, and making the call yourself when needed
• Strong English communication, able to explain complex things to technical and non-technical people alike
• Based in the Helsinki/Espoo region, or open to relocating (we offer relocation support to Finland).

• JavaScript and UI/backend experience, since tooling and APIs touch that layer
• MLOps experience bringing ML/AI systems into production
• Experience with MCP servers and runtimes, including building your own

You do not need to tick every box to apply. If the role looks like a strong match for your skills and direction, we would like to hear from you.

• An open, easy-going startup environment with people you will actually like working with, and a shared drive to build something good

Discover our full range of relocation jobs with comprehensive support packages to help you relocate and settle in your new location.

• A genuinely international team: 10+ nationalities across several countries
• A product you won't find anywhere else, built on a modern data science and engineering stack
• Low hierarchy and a fun ride — we care about kindness, collaboration, and keeping the work enjoyable
• Room to grow on both the technical and business side, with regular demos, knowledge sharing, and internal talks 🍕
• Flexibility to find a work–life balance that fits. We support remote and in-office work, though we would like you in the office for part of the week to build things together, especially early on
• Starting salary of €4,800–6,000 per month with compensation adjusted to experience and skillset
• A solid benefits package: healthcare, lunch and activity benefits, five weeks of paid holiday from day one (pro-rated based on starting date), phone and equipment, and an options program.

If this sounds like your kind of next step, we would love to hear from you.

We review applications and interview on a rolling basis, so apply sooner rather than later — we may move ahead with candidates while the role is still open.

Our process is meant to be transparent and low-stress. It usually looks like:

• A 30-minute intro call with a Sellforte engineer
• One or two interviews: a conversational round and a hands-on technical case with some smaller problem-solving tasks
• A chance to meet people from the team you would be joining

Application deadline: June 24.

Questions about the role? Reach out to Emil at emil@sellforte.com.