Job Description

ABOUT THE ROLE

We are building a new product at the intersection of agentic AI and hard backend engineering. The work runs deep: distributed systems, demanding data problems, and AI engineering that takes large language models beyond demos, into systems you can trust.

As a Junior AI Developer you will build LLM-powered features at the core of it, side by side with senior engineers — agents, retrieval pipelines, and the evaluation tooling that keeps them honest. You will get structured mentoring, real ownership early, and a steep but well-supported learning curve.

WHAT YOU’LL DO

• Implement AI features across our agent, retrieval, and service layers, with code review and mentoring at every step.
• Ship work that is genuinely done: tested, evaluated, documented, and running on our test and production environments.
• Build proofs of concept that explore new models, agent patterns, and product capabilities.
• Contribute to our evaluation harnesses and AI tooling — quality here is a team sport.

WHAT WE EXPECT FROM YOU

• Learn fast: absorb the stack, ask sharp questions, and turn feedback into visible progress.
• Deliver: own well-scoped features end to end and meet your commitments with quality.
• Think in abstractions: our codebase is built on a few deep concepts — mastering them is the job.
• Make AI part of how you work: use agents and AI tooling deliberately when designing, coding, and testing.

WHAT YOU BRING

• Solid Python and working knowledge of Java (or another object-oriented language).
• Genuine hands-on LLM experience: you have built something with prompting, tool calling, RAG, or an agent framework — however small.
• A practical grasp of how LLMs work: tokens, context windows, embeddings, and what they mean for cost and quality.
• Comfort with REST APIs and the fundamentals of client–server systems.
• A testing reflex — you do not consider a feature done without tests and evals.
• Clear communication and the appetite to learn in a demanding environment.

BONUS POINTS

• Agent frameworks (LangChain4j, LangChain, etc.) or MCP.
• Vector stores and embedding pipelines.
• Running or fine-tuning open-weight models locally (Ollama, vLLM, LoRA).
• Exposure to dependency injection (Guice, Spring), Maven, Git workflows, and CI.
• SQL and a clear mental model of how data is modeled and persisted.

QUALIFICATIONS

• Bachelor degree in Computer Science or Computer Engineering.
• Other degrees considered with a proven track record in software development.