An MLOps workflow for training, inference, experiment tracking, model registry, and deployment.

Microservice to digitalize a chess scoresheet

Enterprise Data Warehouse & ML Platform - High-performance platform processing 24B records with <60s latency and 100K records/sec throughput, featuring 32 fact tables, 128 dimensions, and automated ML pipelines achieving 91.2% accuracy. Real-time ML inference serving 300K+ predictions/hour with ensemble models.

scripts for benchmarking vLLM using Llama 8b and NVIDIA 4090 GPU