AI Developer

Full-time · Hybrid (Prague) or Remote · Medior – Senior · English required $4,000 – $8,500 / month + performance bonus

About us

OAKS Consulting has been working in healthcare across Central and Eastern Europe since 2009. We partner with pharma, medtech and healthcare organizations on market access, real-world data design and processing, and the software products we build in-house to support all of the above.

MAiDATA is our healthcare AI data platform. It brings authoritative healthcare datasets together with a modern stack — Snowflake for data, AWS for applications, LLMs for the interface — and serves a portfolio of 50+ pharma and medtech clients across the region.

We ship working systems that change how healthcare data actually gets used.

About the role

You'll own the full lifecycle of AI components: prototype, evaluate, ship to production, monitor, improve. You'll work directly with the SW Development Lead and the data engineering team — no layers between you and the product.

• This is a role for someone who measures success by what's running in production, not by what's on a benchmark.

What you'll work on

• Conversational analytics over MAiDATA — building the next-gen chat interface over our healthcare datasets, with semantic YAML models you'll design and tune.
• MCP (Model Context Protocol) servers — extending our existing MCP layer so MAiDATA data is reachable from Claude Desktop, ChatGPT and our clients' internal tools.
• Smart data ingestion — automated pipelines for new public healthcare datasets, with LLM-assisted classification and quality checks.
• Custom client uploads — secure, tenant-isolated workflows that let clients bring their own data into MAiDATA and analyze it alongside the public datasets.
• Feedback & evaluation pipelines — eval harnesses, prompt regression tests and quality monitoring so the assistant actually improves in production, not just in demos.

Tech stack you'll work with

• Snowflake, Anthropic / OpenAI APIs, MCP, RAG, vector search, function calling / tool use
• Python (primary), SQL (Snowflake, Postgres)
• AWS (ECS, RDS, OpenSearch, Lambda, EventBridge, S3), Terraform, Docker, GitHub Actions with OIDC
• Auth0

You don't need to know all of this on day one. You do need to be the kind of engineer who picks new pieces up fast.

What we're looking for

Must have

• Strong Python and Git.
• Production experience with LLM applications — not just notebooks. RAG, prompts, evals, tool use, agentic patterns.
• Solid SQL and comfort working with relational / warehouse data.
• Experience deploying AI services on AWS.
• Ability to communicate in English with the team.
• Bias toward shipping. We care about working systems in production, not papers.

Nice to have

• Snowflake or another modern data warehouse.
• MCP servers, agent frameworks or LLM orchestration tooling.
• Healthcare or pharma background.
• Classical ML / NLP background (TensorFlow, PyTorch, Hugging Face) for cases where it's the right tool.
• Terraform, ECS, infrastructure-as-code experience.
• Open source contributions or shipped side projects we can look at.
• We weight portfolio over credentials. A bachelor's in CS / engineering is helpful but not required if you can show what you've built.

What we offer

• Competitive base salary (range above) plus performance bonus
• 25 days vacation + 5 sick days
• MultiSport card
• Annual education budget (courses, certifications, conferences)
• Flexible hours, hybrid setup with 2 office days / week at IBC Pobřežní, Praha 8
• Real ownership — small team, senior peers, no layers between you and the product
• Direct exposure to the healthcare data ecosystem