BharatMLStack
BharatMLStack is an open-source, end-to-end machine learning infrastructure stack built at Meesho to support real-time and batch ML workloads at Bharat scale
What is BharatMLStack?
BharatMLStack is an open-source, end-to-end machine learning infrastructure stack built at Meesho to support real-time and batch ML workloads at Bharat scale
Overview of BharatMLStack
What it is
BharatMLStack is an open-source, end-to-end machine learning infrastructure stack built at Meesho to support real-time and batch ML workloads at Bharat scale. It is a production-ready, cloud-agnostic ML infrastructure platform that powers real-time feature serving, model inference, and embedding search at massive scale.
Key features
- Workflow Integration & Productivity: 3x faster experiment-to-deployment cycles and 95% reduction in model onboarding time.
- Cloud-Agnostic & Lock-In Free: Runs across public cloud, on-prem, and edge; Kubernetes-native with zero vendor lock-in.
- Economic Efficiency: 60–70% lower infrastructure costs vs hyperscaler managed services; optimized resource utilization across CPU and GPU workloads.
- Availability & Scalability: 99.99% uptime across clusters; 1M+ QPS with low latency.
- Core Components: Includes TruffleBox UI, Online Feature Store, Inferflow, Numerix, Skye, Go SDK, Python SDK, Interaction Store, and Horizon.
Use cases
BharatMLStack powers personalized candidate generation, personalized ranking, fraud and risk detection, image search, LLM recommender systems, and DL and LLM deployments at scale.
Getting started
bashgit clone https://github.com/Meesho/BharatMLStack.gitcd BharatMLStack/quick-start#Set versionsONFS_VERSION=v1.2.0 HORIZON_VERSION=v1.3.0 TRUFFLEBOX_VERSION=v1.3.0 NUMERIX_VERSION=v1.0.0./start.sh
License and pricing
BharatMLStack is open-source software licensed under the BharatMLStack Business Source License 1.1. Pricing type is open_source.
Based on the project's public README and repository metadata.