Data Scientist (AI-first)

About Sqilline

Sqilline Health builds Danny, an oncology analytics platform that works with real-world clinical data from 120+ hospitals in Bulgaria, Romania, Serbia, Croatia, Spain, and Poland. Pharma companies use our analyses to see how cancer treatments perform in real patients. We run our own LLMs on air-gapped infrastructure, so patient data never leaves our environment.

The role

We are an AI-first company. Our models read raw clinical records and turn them into structured oncology data that some of the biggest pharma companies use to see how cancer treatments perform in real patients. This role is the logic in between: disease-specific algorithms that turn records for an indication into analysis-ready variables and cohorts, and models that pull the right facts out of messy clinical text.

This is a mid-level role with room to grow into owning a disease area. You work on the logic for your assigned indications, with the CDO and the data science team close by. You design the algorithm, build the training set from real corrected cases, and check the result against a gold set before it ships.

The data is patient-level oncology data from hospitals in six countries, so a wrong value distorts an analysis a client acts on. Getting the algorithm right is the job.

You also work AI-first. You build model-driven pipelines around your own workflows: agents do the first pass, you review and judge, the system logs what got corrected. Every algorithm you ship leaves reusable automation behind it.

What you will do

• Deliver analyses on real-world oncology data: cohorts, treatment patterns, outcomes across hospitals, indications and countries.
• Design and implement disease-specific algorithms: the logic that turns clinical records for an indication into analysis-ready variables and cohort definitions.
• Grow into fine-tuning our self-hosted LLMs for clinical extraction: build training sets from corrected cases, run the fine-tunes with support, evaluate against gold sets.
• Verify and validate patient-level data with model assistance: the model does the first pass, flags what it's unsure about, you judge the rest.
• Rebuild recurring delivery workflows as model-driven pipelines, so each month runs more automated than the last.
• Keep your work in proper repositories: versioned code, CI on GitLab, reviewable and repeatable.
• Work with both our stacks: self-hosted vLLM for volume work, frontier models for planning and hard cases.
• Measure everything. Every automated step has an accuracy number; every analysis has a quality trail.
• Show the team how to run and extend what you build.

What we look for

• AI-first thinking. Your default question is "can a model do this?" before you do it by hand even once.
• You use LLMs daily in your own work: coding agents, prompt pipelines, evals.
• You take data quality personally. In clinical data, a wrong value can distort an analysis a client acts on.
• You ship. A working flow this week beats a framework next quarter.
• You measure. An analysis without a quality check, or automation without an accuracy number, is not done.

Skills

You need:

• Solid Python for data work (pandas or similar).
• Strong SQL.
• Statistics fundamentals: distributions, testing, knowing when a result is noise.
• Hands-on LLM work: prompt pipelines, structured output, evaluation, and first experience with fine-tuning (LoRA or similar).
• Git as a habit, Docker and Linux without fear.

Nice to have:

• Survival analysis or other clinical and epidemiological methods.
• Healthcare data experience: EHR, clinical registries, GDPR.
• Airflow, dbt, or similar pipeline tooling.
• Experience with agent frameworks or multi-step LLM workflows.

What we offer

• A challenging role in a dynamic organization
• Competitive remuneration package
• Comprehensive social package, including additional private health insurance and multisport card
• A supportive, friendly, and innovative team environment
• Opportunities for professional growth and development