---
title: Production ML case studies by Sudhanva Narayana
description: Public scope, engineering decisions, tradeoffs, reliability, and measured outcomes.
canonical: https://sudhanva.me/work/
last-updated: 2026-08-21
---

# Production ML case studies by Sudhanva Narayana

Sudhanva Narayana is a Senior Machine Learning Engineer focused on production
ML systems, infrastructure, inference, and model serving. These first-party
case studies describe public scope, technical decisions, operational tradeoffs,
and measured outcomes. Employer-specific implementation details and proprietary
scientific context are intentionally generalized.

## Case studies

- [Billion-Row Batch Inference](https://sudhanva.me/work/case-studies/billion-row-batch-inference/):
  more than one billion rows processed end to end in under three hours.
- [Production ML Platform on Kubernetes](https://sudhanva.me/work/case-studies/production-ml-platform-kubernetes/):
  Ray and Flyte workloads, observability, internal applications, and more than
  $50,000 in annual cloud-cost reduction.
- [Multi-GPU Model Build and Deployment](https://sudhanva.me/work/case-studies/multi-gpu-model-build-deployment/):
  50% shorter deployment time and about ten engineering hours saved each week.
- [Real-Time Transformer Prediction](https://sudhanva.me/work/case-studies/real-time-transformer-prediction/):
  a real-time event pipeline analyzing more than one million daily interactions.
- [Multi-Regional Geospatial ML](https://sudhanva.me/work/case-studies/multi-regional-geospatial-ml/):
  more than 1 TB daily, 75% better inference efficiency, and 50% lower
  map-rendering latency.

The [human-readable work index](https://sudhanva.me/work/) also includes
open-source engineering projects and links to relevant technical writing.