Senior Data Scientist, Search & Matching Platform

We are hiring a Senior Data Scientist for our client - a fast-growing, VC-backed technology startup building a data-intensive AI product.

Our client is an international startup that achieved 18× growth in 2025, placing it in the top <1% of startups globally.

The company is building a live AI-powered job matching platform with 50k+ monthly active users.

Data is the company’s core asset, and the next stage of growth depends on how effectively search, retrieval, ranking, and semantic understanding systems are designed and improved in production.

As part of this phase, we are looking for a Senior Data Scientist who will take ownership of search and matching models end-to-end - from raw data to production impact.

• Audit existing data sources and build training-ready datasets for ML models

• Work with heterogeneous data sources, including:

  • ATS and job descriptions

  • Application forms and screening questions

  • Candidate profiles

  • User behavior and event logs

• Define data requirements together with engineers:

  • logging standards

  • data formats

  • metrics

  • monitoring signals

• Design and implement feature engineering pipelines:

  • from unstructured text (job descriptions, applications)

  • from structured fields

  • from behavioral and interaction logs

• Build reusable, well-documented features suitable for both offline training and online inference

• Identify new signals, insights, and data sources that improve matching quality

• Design, build, and iterate on retrieval and ranking systems, including:

  • Hybrid search (sparse + dense):

    • BM25 / TF-IDF combined with embeddings

    • weight tuning, boosting strategies, recall evaluation

  • Cold-start (non-personalized) rankers followed by personalization based on user behavior

• Build:

  • strong baselines

  • Learning-to-Rank (LTR) models

  • classification + ranking pipelines where appropriate

• Own offline and online evaluation, including:

  • metric design

  • experimentation

  • production monitoring of model performance

• Use LLMs for structured information extraction from unstructured text

  • prompt design for structured (e.g., JSON) outputs

  • schema validation and automated repair

  • robustness to noisy or inconsistent inputs

• Define and track quality metrics:

  • field-level accuracy

  • unknown or missing field rates

  • extraction stability over time

• Work closely with backend and ML engineers to:

  • deploy models to production (batch and near real-time)

  • define APIs and inference flows

  • ensure observability and reliability

• Take responsibility for the full lifecycle:

  • from data and features

  • to models

  • to measurable production impact

• Deep understanding of existing data sources and the search pipeline

• A production-ready baseline retrieval and ranking model

• Clear metrics and monitoring for search and matching quality

• Tangible improvements in recall and ranking relevance

• 3+ years of experience in Data Science or Machine Learning

• Strong background in search, ranking, recommender systems, or closely related problems

• Proven experience bringing ML models into production environments

• Strong Python skills

• Solid SQL knowledge

• Hands-on experience with:

  • ranking or Learning-to-Rank models

  • feature engineering for text and structured data

  • offline and online model evaluation

• Experience with deep learning and embeddings

• Experience with:

  • search engines (Elasticsearch / OpenSearch)

  • vector databases

  • ANN indexes

• Familiarity with HR tech, ATS systems, or marketplace products

• Practical experience using LLMs in production systems

An example of the type of problem the team works on:

Building a Skills Taxonomy (Skills Graph)
Transform raw job requirements into a structured skill representation with taxonomy placement and weighted relationships between skills (e.g., similarity, co-occurrence, domain grouping).

• Market-level compensation

• Office / hybrid in Warsaw or fully remote within nearby time zones (CET ±2)

• 20 paid working days of vacation per year plus sick leave

We look forward to receiving your CV and learning more about your experience!
Dear Candidates, due to a high volume of applications, only selected candidates will be contacted for interviews. We appreciate your understanding. Thank you for considering a career with us.

• Work type
• Full-time
• Location
• Warsaw