SalesRook
Senior Technical Lead · 2024 - Present
Lead generation and qualification platform for letting agents. Handles enquiries on autopilot.
LiveIn production for the UK property market
#saas#ai#automation
AI EngineerTechnical LeadRAG & AgentsFull-StackMobile
Shanker Dhand
EngineeringAIIntoProduction
I lead a small engineering team shipping AI agents, RAG pipelines, and automation into production — built on 10+ years of full-stack and mobile engineering.
Selected work
Products and platforms I've shipped.
Distribution Management
Senior Software Engineer · 2018 - 2024
Insurance distribution platform for producer onboarding, licensing, and compensation across enterprise carriers.
EnterpriseUsed by insurance carriers across the US
#saas#enterprise#insurance
Barbr Biz
Mobile & Backend Engineer
Booking and business platform for barbers — native iOS and Android apps plus the backend architecture powering scheduling, payments, and client management.
ShippedNative apps live on the App Store and Google Play
#mobile#saas#booking
Kami Driving School
Mobile Engineer & Product Designer
Business platform for a driving school — public booking site plus internal and public mobile apps to manage lessons, scheduling, and day-to-day operations. Also designed the product and wrote the site's content.
End-to-endProduct design, apps, backend, and site content
#mobile#booking
Twindow
Creator & Engineer
Free, native multi-instance launcher for macOS (Windows soon) — clones any app into an isolated copy with its own account, data, and Dock identity. Built for the community.
FreeIndependent macOS product, designed and built solo
#macos#tools
What I do
Ways I can help.
The stack
About
Technical Lead & AI Engineer.
10+ years in software — I started in mobile (native Android, then React Native), scaled into full-stack engineering, and spent nearly six years at Duck Creek Technologies building enterprise insurance platforms used by carriers across the US.
- 10+ years in software
- 6+ years at an MNC (Duck Creek Technologies)
- Full-stack, mobile & AI
- AI agents & automation in production
My journey
From native mobile to AI in production.
Foundations
Today
- Native AndroidWhere it began · 2015 - 2016
Learning to ship
First real apps in native Android — real screens, real bugs, real users. Learned what it takes to put software in people's hands: offline states, performance on cheap devices, release discipline.
- iOS & SwiftA second platform · 2016 - 2017
Thinking beyond one platform
Built native iOS apps alongside Android. Learned that platforms differ but products don't — and started designing systems that outlive any single stack.
- React NativeOne codebase, every device · 2017 - 2018
From apps to products
Shipping both platforms from a single codebase. Learned to think in components and shared logic — the first step from building apps toward building the systems behind them.
- React.js & Node.jsEnterprise SaaS · 2018 - 2024
Full-stack at enterprise scale
Six years on insurance platforms used by carriers across the US — first the web frontend in React, then deep into the backend with Node and TypeScript. Learned how enterprise SaaS is really built: multi-tenant scale, compliance, and software that businesses run on.
- AI & AutomationThe current chapter · 2024 - Now
Engineering AI into production
RAG pipelines, AI agents, and CRM automation running a live SaaS platform — leading a team of five. Everything underneath — mobile, full-stack, enterprise — became the foundation for AI systems that actually ship.
Writing
Latest writing.
- ·11 min read
MCP's 2026-07-28 spec broke my tool-calling setup
Model Context Protocol's 2026-07-28 spec ships a stateless core, drops Roots, Sampling, and Logging, and hardens auth to OAuth 2.0 — what to migrate first.
- ·5 min read
Chunking strategies for RAG, and when each one bites you
Fixed-size, semantic, and structure-aware chunking all look fine in a demo. Here's how each one actually fails once real documents hit your production pipeline.
Working on AI?
Tell me what you're building. I read everything and reply within 48 hours.