Data Scientist

Role & responsibilities

We are looking for a hands-on Data Scientist specializing in Natural Language Processing to design, build, evaluate, and deploy production-grade NLP and machine-learning solutions for complex text-driven workflows.

Core Responsibilities

• Design and build NLP and machine-learning pipelines that transform noisy, heterogeneous text data into clean semantic layers ready for modeling, retrieval, analytics, and downstream product use.

• Develop retrieval systems for semantic search, candidate ranking, out-of-vocabulary handling, and high-quality information discovery using embeddings and vector search.

• Build, compare, and evaluate supervised and hybrid approaches, including multioutput classifiers, hierarchical classifiers, NER-style parsers, clustering techniques, and rule-plus-ML systems.

• Analyze decision boundaries and relationships across free-text fields using conditional distributions, entropy, mutual information, directional association, embeddings, and predictive ablation studies.

• Deploy, monitor, and continuously improve ML services in collaboration with platform, backend, data engineering, and product teams.

Required Skills

• Strong Python and SQL skills, with hands-on experience building production data pipelines and machine-learning workflows.

• Practical experience with text classification, semantic similarity, embeddings, information retrieval, ranking, clustering, or entity extraction.

• Experience with Hugging Face, SentenceTransformers, tokenization, fine-tuning, transformer-based models, and model evaluation. • Solid understanding of vector search, cosine similarity, approximate nearest-neighbor methods, and retrieval metrics such as Recall@K, MRR, and NDCG.

• Ability to design experiments, define evaluation metrics, compare model trade-offs, and communicate findings clearly to technical and non-technical stakeholders. Nice to Haves

• Experience with healthcare, imaging, enterprise metadata, document intelligence, search, recommendation, knowledge retrieval, or routing systems.

• Knowledge of DICOM, PACS/RIS, HL7/FHIR, or other structured-plus-free-text enterprise data standards.

• Familiarity with MLOps, cloud deployment, API-based model serving, monitoring, responsible AI practices, or privacy-aware handling of sensitive text data.