Join our team

At Bloomteq, we know that exceptional organizations are built by exceptional people. That’s why we are always open to connecting with individuals who are curious, motivated, and eager to grow. We operate at the intersection of technology, innovation, and research, in an environment where collaboration, ownership, and continuous learning are part of everyday work. If you are passionate about creating meaningful solutions, developing your skills, and working alongside people who support and challenge each other, you might be a great fit for our team. Submitting an open application is an opportunity to introduce yourself, share your experience and interests, and tell us how you see your potential contribution. When a suitable opportunity arises, we will be happy to reconnect with you.

About the role

You'll build agentic AI systems that do real work—not demos. That means designing agents that plan, call tools, handle failures, and produce output a business can actually depend on.

We're moving toward AI-centric engineering across the company, and this role sits at the centre of it. You'll have real influence over how we build with these tools, what we standardize on, and where the line sits between what an agent should handle and what still needs a human.

What you'll work on

• Designing, building, and shipping agentic systems and LLM-powered workflows in Python.
• Working with LangChain and similar frameworks to orchestrate multi-step agent behaviour.
• Prompt design, tool and function calling, and retrieval pipelines using RAG.
• Evaluating agent performance by building the tests and metrics that determine whether it actually works.
• Integrating models and agents into production services and existing product workflows.
• Properly handling the unglamorous parts: cost, latency, failure modes, guardrails, and observability.
• Working with engineering and product teams to identify where this technology genuinely helps—and saying so when it doesn't.

What you need

• 3+ years of software development experience with Python.
• Hands-on experience building with LLMs and agentic frameworks such as LangChain or an equivalent—with something that reached production.
• An understanding of RAG, embeddings, vector stores such as Qdrant or Pinecone, and retrieval design.
• Experience treating prompt engineering and tool calling as engineering problems rather than experiments.
• The ability to evaluate LLM output systematically rather than by impression.
• Comfort integrating with REST APIs and cloud services.
• Strong ownership, communication, and problem-solving skills.

Strong advantages

Experience with any of the following is a strong advantage:

• LangGraph
• LlamaIndex
• Vector databases such as Qdrant, Pinecone, Weaviate, or pgvector
• AWS services such as Bedrock, Lambda, or SageMaker
• MLOps and model deployment
• A classical machine-learning background
• Fintech or payments
• Docker

Don't meet every requirement? Apply anyway. Nobody in this field does.

What matters most to us

We're not looking for someone who's built a chatbot demo.

We want an engineer who understands where these systems break, builds the evaluation that catches those failures, and can tell the difference between an agent that looks impressive and one that can be trusted in production.