Case StudyEnterprise AIPortfolio

Production AI Lessons from Four Documented Projects

By Anurag Srivastav3 min read

A behind-the-scenes look at documented AI and automation work involving HP, Intel and BMRC Hospital during prior employed roles — and what separates production systems from demos.

Neo-brutalist engineering workbench featuring Linux container deployments, Intel & HP telemetry monitors, and automated workflow daemons
Verified Research · Oct 2026

Behind four production client engagements: from enterprise Docker workbenches to multilingual hospital triage.

01

Anyone can build an AI demo. Shipping AI that an enterprise depends on every day is a different discipline. This case study walks through the real systems I have delivered as an AI Automation Engineer in India.

02

Gignaati Workbench for HP and Intel

I architected the Gignaati Workbench during a prior employed role for a project involving HP and Intel, plus a companion Docker automation desktop app. The portfolio reports an 85% deployment-time reduction; the underlying benchmark is not independently published.

03

BMRC Hospital AI Feedback Agent

For BMRC Hospital I built a multilingual AI feedback agent that processes patient feedback, runs sentiment analysis and routes tickets automatically. In healthcare, auditability and language coverage matter more than raw model size.

04

What separates production from demos

Observability, idempotency, fallbacks, and measurable outcomes: 200+ automated operations a day, 85% faster deployment, and a 10-day AI training program delivered to 200+ PhD professors. These are reported portfolio outcomes. The linked project case studies describe their scope and identify the measurements that are not publicly available.

Primary references

Explore the related service →

Need help with an automation or LLM workflow?

I build n8n workflows, LLM applications and Python integrations for bounded operational problems, with evidence limits stated on the related case studies.

Get in touch