Senior Full Stack Engineer · Solutions Architect · Founder of Genzy Nanny
Washington, DC · Telugu · Hindi · English

I’m a Senior Engineer and Architect with 15+ years building scalable, secure platforms — federal cloud services, Backstage developer portals, agentic AI systems, and now a live consumer marketplace.

I care about products that remove friction from real life. The hardest problems I take on sit at the intersection of trust, payments, identity, and matching.

Daycare waitlists fill months out. A closed school day still lands on a parent’s calendar. Plenty of sitters live on the same street and want fair, flexible work — and families still cannot meet them with proof they can trust.

That gap is why I built Genzy Nanny.

Not another classifieds board. A system where identity, availability, booking, chat, and pay stay in one place — so parents can look someone up before they knock, and sitters get booked by families who already trust the process.

Help is often next door. Trust should be too.

Genzy Nanny is a USA childcare marketplace I founded and built end to end: web, iOS, backend, payments, identity, and ops.

Parents find verified sitters nearby. Sitters set rates and get paid in the app. Instant Booking alerts eligible caregivers in waves until the first accept wins.

Seller: Sasi Kesani · Company: GenzyNanny Inc. · Latest: iOS 1.0.4 · Rating: 5.0 on the App Store

• Government ID + facial match (Didit)
• Background check review
• CPR / First Aid badges
• Public names as first name + last initial until a booking exists
• Secure in-app messaging and Stripe-backed escrow-style payments

1. Parent posts time, place, kids, and budget
2. Nearby opted-in sitters are ranked by proximity and trust
3. Alerts go out in stages (immediate, then expanding waves)
4. First successful accept gets the job
5. Card authorizes up front and captures only when a sitter is confirmed

React Native · Expo (web + iOS) · TypeScript · Supabase (Postgres, Auth, RLS, Realtime, Edge Functions) · Stripe (PaymentIntents + Connect) · AWS S3 + Cloudflare · Terraform · GitHub Actions · Didit KYC · Resend / Brevo · OpenRouter

I live in the Washington, DC area. I have spent most of my career as the person teams call when the architecture is messy and the deadline is real — Java and Spring Boot services, Angular and React front ends, AWS and Azure platforms, Kubernetes, and now agentic AI on Bedrock, LangGraph, and MCP.

Colleagues I have worked with on federal / GSA-IT programs have described me as the go-to for hard problems: architect the framework, ship the feature, mentor the junior engineer, and stay until it works.

What I am unusually good at

• Turning a fuzzy product into a working marketplace (auth, search, bookings, money, notifications)
• Trust boundaries: RLS, identity verification, least-privilege admin, no client-side money math
• Developer platforms: Backstage catalogs, Entra ID SSO, GitHub sync, TechDocs
• Agentic workflows that are engineered, not just prompted — typed tools, HITL, retries, MCP

Languages I speak: Telugu (native) · Hindi · English

AI-driven, bio-inspired cybersecurity for microservices — autonomous threat detection and polymorphic defense. Java / Python / Node SDKs, Kubernetes-ready. npm: sdi-nodejs

An engineered peer-review agent: LangGraph + MCP, Pydantic structured output, human-in-the-loop on every tool. Built to show how agentic AI belongs in a real software workflow, not as a one-shot prompt.

Modular service catalogs with Microsoft Entra ID SSO, GitHub sync, and TechDocs automation — so teams can see what they own.

ATS-oriented resume drafting from natural-language prompts and job alignment.

Cross-cloud DevSecOps with Terraform, Kubernetes, LangChain, and AWS Bedrock for orchestration and assistants.

A curated collection of real-world LLM apps, agents, and RAG patterns.

• Trust-first consumer marketplaces (identity, matching, escrow, staged dispatch)
• Agentic AI on AWS Bedrock, LangGraph, and MCP — including code-review and DevOps assistants
• Backstage plugins for catalog intelligence and service visibility
• Cloud cost and privacy-aware GPU / inference runtimes

⭐ If you care about childcare, trust systems, or agentic engineering — connect. I am always happy to talk shop or collaborate.