Data Scientist

Job Title: Data Scientist Location: Bangalore (Hybrid/On-site depending on project needs)About the RoleWe are seeking a highly skilled Data Scientist to join our team in Bangalore. In this role, you will take ownership of data science components across client projects, build production-ready ML and GenAI-powered applications, and mentor junior team members. You will collaborate with engineering teams to design and deploy impactful solutions that leverage cutting-edge machine learning and large language model technologies.Key ResponsibilitiesML & Data ScienceDevelop, fine-tune, and evaluate ML models (classification, regression, clustering, recommendation systems).Conduct exploratory data analysis, preprocessing, and feature engineering.Ensure model reproducibility, scalability, and alignment with business objectives.GenAI & LLM ApplicationsPrototype and design solutions leveraging LLMs (OpenAI, Claude, Mistral, Llama).Build RAG (Retrieval-Augmented Generation) pipelines, prompt templates, and evaluation frameworks.Integrate LLMs with APIs and vector databases (Pinecone, FAISS, Weaviate).Product & Engineering CollaborationPartner with engineering teams to productionize ML/GenAI models.Contribute to API development, data pipelines, technical documentation, and client presentations.Team & GrowthMentor junior data scientists and review technical contributions.Stay up to date with the latest ML & GenAI research and tools; share insights across the team.Required Skills & Qualifications4.5–9 years of applied data science experience.Strong proficiency in Python and ML libraries (scikit-learn, XGBoost, LightGBM).Hands-on experience with LLM APIs (OpenAI, Cohere, Claude) and frameworks (LangChain, LlamaIndex).Strong SQL, data wrangling, and analysis skills (pandas, NumPy).Experience working with APIs, Git, and cloud platforms (AWS/GCP).Good-to-HaveDeployment experience with FastAPI, Docker, or serverless frameworks.Familiarity with MLOps tools (MLflow, DVC).Experience working with embeddings, vector databases, and similarity search.