Job Role: Machine Learning Engineer III
Location: Bangalore, Karnataka
Experience: 5-8 Years
Position Overview
Join Swiggy's Data Science Platform (DSP) team as a builder of the foundational systems that enable our Data Science teams to move from experiment to production. You will work at the intersection of applied ML and platform engineering - partnering with Data Scientists to turn research into reliable, scalable production systems. The role demands hands-on fluency across the full ML stack: modeling (classical, deep learning, and generative AI), productionization, platform infrastructure, and operational excellence.
What you'll get to do here:
. DS Collaboration & Code Review: Partner with Data Scientists to review notebooks and training scripts, and translate research prototypes into production-grade code. . Model Engineering & Productionization: Own notebook-to-production for classical ML, deep learning, and generative AI models - covering training pipeline optimization, ONNX export, quantization, and serving integration. . Platform Engineering: Build and maintain reusable ML platform components - feature stores, model registries, serving infrastructure, experimentation platforms, and model governance frameworks. . Data Pipelines & Scalable Infrastructure: Build and optimize batch and real-time ML pipelines on Databricks and Snowflake, with end-to-end automation for training, validation, and deployment. . MLOps, Observability & Opex: Run CI/CD for ML models, own data drift and model quality monitoring, and take full opex responsibility - cost, incidents, capacity, and SLAs. . Generative AI & LLM Integration: Integrate LLMs, embeddings, and RAG pipelines into the platform, and manage LLM serving infrastructure for cost, rate limiting, and latency at scale.
What qualities are we looking for
. ML & AI Depth: Strong grasp of ML fundamentals (supervised/unsupervised learning, regularization, validation) across model families - tree-based, neural networks, transformers, and embeddings. Familiarity with GenAI architectures and best practices in training and inference. Exposure to LLMs, fine-tuning, or prompt engineering is a plus. . Engineering Excellence: Expert Python skills with a track record of writing clean, reproducible, production-grade code. Proficiency in TensorFlow or PyTorch for training and serving. Familiarity with containerization (Kubernetes/Docker) and cloud-native ML services. . Platform & Data Systems: Hands-on with Databricks (jobs, Delta tables, workflows, cluster sizing and cost trade-offs) and Snowflake. Proven experience building ETL pipelines using Spark (PySpark/Scala), Hive, or Presto. Working knowledge of stream processing (Kafka/Flink), feature stores, and experiment tracking tools (MLflow, Weights & Biases, or similar). . Production, Serving & Scale: Experience deploying and operating models in production via REST/gRPC serving, with a clear understanding of latency budgets and SLA management. Ability to diagnose and resolve performance bottlenecks - covering quantization, batching, caching, async inference, and horizontal scaling. . MLOps & Reliability: Practical experience with data drift detection, model observability, and experimentation platforms (XP/A/B testing). Versed in model governance, lifecycle management, and feature reusability. Treats the ML platform as a product - owns cost optimization, incident response, and capacity planning, not just deployment. . DS Partnership, Domain & Experience: 5+ years of experience, with at least 3 in ML Engineering, MLOps, or applied Data Science. Proven ability to work alongside Data Scientists - reviewing modeling decisions and co-owning model quality end-to-end. Familiarity with at least one applied ML domain (recommendations, search, pricing, demand forecasting, or operations research) is strongly preferred.
Bonus Points If You Have:
. Experience with LLM orchestration frameworks (LangChain, LlamaIndex) or fine-tuning open-source models. . Familiarity with vector search systems (Pinecone, Weaviate, pgvector, or similar). . Contributions to internal ML platforms, developer tooling, or open-source ML/AI projects.
We are an equal opportunity employer and all qualified applications will receive consideration for employment without regard to race, colour, religion, sex, disability status, or any other characteristic protected by the law
