Data Scientist

• Fine-tune transformer models and LLMs using techniques such as LoRA and QLoRA for entity extraction, classification, and reranking.

• Fine-tune transformer models and LLMs using techniques such as LoRA and QLoRA for entity extraction, classification, and reranking.
• Develop NLP pipelines to extract structured entities, skills, and relevant information from resumes, job descriptions, and other unstructured datasets.
• Build predictive models for entity classification, entity linking, and skills intelligence.
• Design and implement knowledge graphs, graph traversal methods, and relationships across large skills taxonomies.
• Develop retrieval, clustering, semantic search, and learning-to-rank pipelines using embeddings, bi-encoders, and cross-encoders.
• Apply classical machine learning techniques, including gradient boosting, logistic regression, clustering, and dimensionality reduction.
• Design robust evaluation frameworks, analyse model errors, select appropriate metrics, and validate improvements through statistical analysis.
• Process and analyse large datasets containing millions of records using Python, SQL, and relevant data processing tools.
• Deploy and optimise machine learning models for production environments, ensuring scalability, reliability, and performance.
• Use experiment tracking, model versioning, Docker, and AWS-based GPU training and serving workflows.
• Collaborate with engineering and product teams to integrate AI capabilities into enterprise applications.
• Document experiments, communicate findings, and continuously improve model quality, efficiency, and business impact.

• 1–3 years of experience for Data Scientist I or 3–5 years for Data Scientist II in applied NLP, machine learning, or a related field.
• Strong proficiency in Python and practical experience with PyTorch, Hugging Face Transformers, sentence-transformers, pandas, and scikit-learn.
• Hands-on experience fine-tuning transformer models on custom datasets for named entity recognition (NER), classification, or LLM-based tasks using PEFT techniques such as LoRA/QLoRA.
• Strong understanding of embeddings, attention mechanisms, contrastive learning, semantic retrieval, bi-encoders, and cross-encoders.
• Working knowledge of classical machine learning algorithms and statistical evaluation methods.
• Experience with SQL and analysing large-scale datasets.
• Understanding of learning-to-rank systems, knowledge graphs, entity linking, or large taxonomies.
• Familiarity with ML lifecycle practices, including experiment tracking, model versioning, Docker, and cloud-based model deployment.
• Strong analytical thinking, problem-solving ability, and attention to model quality.
• Clear written communication skills and the ability to explain experiments, results, and technical trade-offs.