micro-batching

Ranabhum is a highly scalable, event-driven benchmark and real-time telemetry pipeline designed to compile, isolate, stress-test, and score high-performance financial order matching engines submitted by contestants.

Production-grade AI security middleware with async micro-batching, prompt injection detection, RBAC enforcement, PII redaction, cost tracking, and concurrency benchmarking.

Production-grade ML model serving platform — versioned registry, async micro-batching, canary routing, prediction logging, and PSI drift detection. FastAPI · PostgreSQL · Redis · Docker · Kubernetes.

Inference gateway for batching AI model requests through bounded micro-batching, queue backpressure, per-request result routing, and reproducible performance evaluation.

Benchmark harness that turns model-serving configs into dollars per 1M requests. Serves a 4.9M-param intent classifier as fp32, dynamic int8, and TorchScript variants with and without micro-batching, measures p50/p95/p99, req/s, and peak RSS, and prices throughput against a dated AWS snapshot.