Shivam Govind Rao

Full-stack engineer with 2.5+ years shipping production React frontends and Node.js/Python backends end-to-end — component architecture and state management through REST API design, distributed queues, and AWS infrastructure. Sole engineer on two AI products that both reached paying customers. Comfortable owning features from requirement through implementation, testing, and delivery.

Software Development Engineer at Diagna AI (FAXFlo) (2025-01 – 2026-04)

First Engineer · Full-Stack

• Shipped the entire product solo: React 19 frontend (multi-tenant architecture with per-clinic feature flags, RBAC routing, ∼60% bundle reduction via code splitting) and production Node.js/Express/PostgreSQL backend (40+ REST endpoints, JWT/RBAC, Swagger) — took FAXFlo from company pivot to first paying US clinic in under 6 months.
• Architected a distributed AWS document pipeline (S3 · SQS · Textract · SNS · Bedrock) processing 1,000+ medical documents daily at 99%+ uptime — BullMQ/Redis workers with dead-letter queue, idempotency keys, and a scheduled auto-retry sweep that self-heals transient failures without manual intervention.
• Integrated a multi-model AI chain — Claude Sonnet via AWS Bedrock, GPT-4o cross-check, and Amazon Nova Pro retry fallback — for document classification at ∼95% accuracy across 40+ medical categories — custom prompting, structured JSON extraction, confidence scoring, and retry logic for malformed model outputs.
• Built a two-stage image compression layer with sharp: per-image iterative resize/quality reduction to fit Bedrock's 5 MB image limit, and batch-level Lanczos3 downscaling on retry when multi-page payloads exceeded the combined cap — cutting payload sizes by up to ∼80%.
• Engineered a Python RPA suite (Robocorp · Playwright · FastAPI) to write structured data into an EHR with no usable API — OTP/2FA login, patient search with middle-name disambiguation, and document upload via browser automation — plus an AI SMS orchestrator across Twilio, Telnyx, and Vonage with provider failover.
• Added Jest + Supertest integration tests on critical document paths, Swagger/OpenAPI docs across all endpoints, and a HIPAA-compliant PostHog analytics dashboard — the interface used in the investor demo that closed the first paying clinic.

Frontend Developer at Oriserve (VoiceGenie.ai) (2023-12 – 2025-01)

Sole Frontend Engineer

• Built VoiceGenie from scratch in 11 months as the sole frontend engineer — customer dashboard, AI campaign builder, post-call analytics, onboarding UX, and marketing site; the product grew from 0 to $10K MRR on what I shipped.
• Built the campaign flow: live @token script composer resolving per-contact variables at call time, ElevenLabs voice configuration, real-time call status, and post-call review with emotion-tagged transcripts and AI-extracted entities.
• Cut page load times 30–50% via code splitting, lazy loading, and caching (Lighthouse-profiled); integrated HubSpot, GoHighLevel, and Cal.com end-to-end in TypeScript with no backend engineer handing off contracts.

B.Tech in Computer Science & Engineering – IET, Dr. Ram Manohar Lohia Awadh University, Ayodhya (2019 – 2023)