Data Scientist (AI_ML & Analytics) Quloi

3.0 - 5.0 years

10 - 13 Lacs

pune, mumbai (all areas)

Work from Office

Responsibilities/ Duties •Product collaboration: Translate product needs into clear problem statements, evaluation plans, and straightforward solution flows; contribute to UI/UX considerations for model inputs/outputs and edge cases. •Applied ML: Build and iterate on models for logistics use cases (e.g., chatbot, forecasting, ETA, anomaly detection, recommendations, document parsing), favoring simple, reliable approaches first. •Data preparation: Explore and prepare data from MongoDB Atlas, Kafka/MSK streams, and S3; implement reproducible feature pipelines and basic data quality checks. •Evaluation & rollout: Define metrics and baselines; run offline evaluations and support phased/A-B rollouts with clear success criteria. •Integration: Package models as small services or batch jobs; collaborate with Backend/Frontend to design APIs/GraphQL contracts, payloads, and error handling that fit our core platform. •DevOps collaboration: Request/provision suitable compute (EC2; occasional GPU), containerize jobs (Docker), set up light scheduling (EventBridge/Step Functions or Airflow if used), and add basic monitoring/alerts. JD Data Scientist (AI_ML & Analytics) •Analytics & insights: Perform focused analyses to support product decisions and present findings with clear visuals and concise summaries. •Quality & governance: Follow coding standards; version datasets/models where practical; respect access controls for customer data; keep documentation current (problem briefs, data lineage, model cards, runbooks). Criteria for the Role! •3–5 years in applied machine learning with at least one end-to-end production delivery. •Strong Python (pandas, numpy, scikit-learn; plus XGBoost/LightGBM). •Comfortable with SQL and working with MongoDB/event data (Kafka/MSK). •Experience deploying models as APIs or batch jobs (Docker; Git-based workflows); familiarity with CI/CD concepts. •Understanding of evaluation design for classification/regression/time-series; pragmatic feature engineering. •Clear communication and a collaborative, documentation-friendly approach.

Posted 14 hours ago

Apply

Start Your Job Search Today

Browse through a variety of job opportunities tailored to your skills and preferences. Filter by location, experience, salary, and more to find your perfect fit.