IntelligentDX
Apr 2025 – present
Healthtech software for eye-care practices, Pune, India
Stack at work: Python (Flask), Node.js (Express), C#/.NET, PostgreSQL, MongoDB, Redis, BigQuery, Google Cloud, React/TypeScript
Junior Software Developer
Nov 2025 – present
Full-time
- Built a treatment-plan booking API that books every visit of a multi-appointment plan across up to 15 parallel lanes, with atomic slot reservation, one audit row per appointment and 200/207/422 responses. In production since September 2026.
- Designed background sync (Cloud Scheduler to Cloud Run Jobs to PostgreSQL) for 213 providers × 111 locations × 70 visit types, replacing slow or timing-out external calls with indexed database reads.
- Cut the slot-search API from about 2.7–3.6 s to 1.7–2.5 s with a concurrent pager, and moved scheduling APIs to an EHR's FHIR R4 API with unchanged API contracts.
- Designed an AI rule-generation platform around Claude (architecture, low-level design, API contract) with validation gates and a repair loop. Prompt caching cut cost per generated rule by about 27–31%.
- Built cancel and reschedule APIs on an EHR with no native reschedule: book the new slot first, retry at timed offsets, return HTTP 207 on partial success and keep a full audit trail.
- Owned an accuracy dashboard end to end (Flask and React/TypeScript). QA reconciliation against the database passed 228 of 229 checks.
- Authored 400+ pull requests, reviewed 180+, and wrote 59 design, API-contract and runbook documents.
Software Engineer Intern
Apr 2025 – Oct 2025
Internship, converted to full-time
- Migrated 10 API read paths across 5 C#/.NET microservices from MongoDB to PostgreSQL in about 8 weeks (PL/pgSQL functions, Dapper repository layer, feature-flag fallback). 9 reached production with identical responses.
- Built SSO login, token-validation and token-refresh APIs in a .NET identity service using Microsoft Azure AD.
- Built the first version of a claim-scrubbing integration in Node.js: an XML-to-JSON claim gateway and a Redis-first cache with PostgreSQL and Firestore fallbacks.