MLOps Engineer
HybridBengaluru, Karnataka, India
Job Summary
Deploy, version, and manage ML model endpoints for LLM and computer vision pipelines across multi-region AWS infrastructure. Build CI/CD pipelines for model delivery with automated testing, staged rollouts, and safe rollback procedures. Monitor system health by tracking performance metrics, detecting drift, and leading incident response. Optimize cloud infrastructure costs on SageMaker, Bedrock, and Lambda while maintaining reproducible environments through infrastructure-as-code. Manage experiment tracking and model/data versioning to ensure clean lineage across training and deployment stages. Own reliability, scalability, and observability for the Scam Decipherers team's real-time scam detection platform operating at sub-second latency.
Required Qualifications
- 4–5 years of hands-on experience in software engineering, data engineering, or ML engineering, with meaningful exposure to production systems beyond notebooks or coursework
- Solid end-to-end understanding of ML workflows — data preparation, training, validation, deployment, and monitoring — and the real-world challenges of keeping models reliable in production (latency, versioning, rollback, drift)
- Hands-on AWS experience, including core services (S3, SQS, Lambda) and ideally SageMaker and Bedrock; comfort with containerisation and infrastructure-as-code
- Strong Python skills, solid version control practices, and a working understanding of CI/CD concepts and production observability (logs, metrics, traces, dashboards)
- Full professional proficiency in English
- This position is based in Bengaluru, India, and offers hybrid working flexibility
Desired Qualifications
- Hands-on experience with experiment tracking and data/model versioning tools in production, or with batch and real-time inference pipelines on managed or container platforms
- Exposure to GPU-based model serving, LLM inference optimisation (quantisation, speculative decoding), or multi-region AWS deployments
- Familiarity with cybersecurity domains — scam detection, fraud, or threat intelligence — or experience fine-tuning open-source language or vision models
- Background in data privacy and secure handling of sensitive data in ML systems
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.