Atul Patil

I'm a data engineer with 2 years of experience building production data infrastructure and applied AI systems for enterprise clients, including Fortune 500 organizations and the founder of Acadzen AI, an independent SaaS venture I built and shipped entirely on my own.

At Xcaliber Infotech I engineered a metadata driven ETL platform that expanded source integrations from 1 to 7+ types, processing millions of records across 100+ daily pipeline runs and cutting data load time by 64%. I built a reusable API ingestion framework handling multiple authentication models and deeply nested JSON, along with idempotent loading logic that eliminates duplicate or lost records across retries and restarts. I also automated a manual reporting workflow for a separate initiative, replacing 5 scripts with a single wrapper API call in about a week.

More recently I've been building AI generated insight modules for procurement analytics, covering supplier rationalization, off contract spend, tail spend optimization and supplier fragmentation. That work combines deterministic business metrics with LLM generated narratives, backed by a guardrail layer that validates every AI recommendation against verified data rather than trusting the model output. I also traced and fixed a data reconciliation issue between a declared reporting source and live transactional data that had been quietly invalidating downstream savings calculations.

Outside of work I founded Acadzen AI and built a suite of 15 AI powered SaaS products, owning architecture, backend, React frontend and deployment for every one. The flagship is CompeteIQ, a live competitive intelligence platform that turns a company name into a full analyst grade report in under two minutes, with a chatbot grounded strictly in verified data so it won't hallucinate answers.

I work well async, I'm comfortable owning ambiguous problems end to end, and I'm available for 4+ hours of daily overlap with US and European time zones.