Malwinder Singh

Mohali, India — open to leadership & consulting work

Malwinder Singh

Tech Lead · Agentic AI, Flutter & Web

Tech Lead with 11+ years building mobile and web applications. Today I lead delivery at Netsmartz and build agentic AI web and Flutter apps for clients, working AI-first with Claude and GitHub Copilot. Earlier in my career I delivered apps for OMRON and Emerson, and I run product and the software platform at SPDrone.tech, my family’s drone company.

11+Years in the craft
1M+Stack Overflow reach
50K+Medium article views
10+Products shipped

01 About

Eleven years of engineering judgement, now moving at AI speed.

I’m a results-oriented Tech Lead with over a decade — 11+ years — across mobile development (Flutter, Android, iOS), web applications, and AI/ML solutions. Much of that time was spent close to the metal: BLE and NFC stacks, IPC, on-device ML, and the kind of root-cause analysis that happens when a device in the field misbehaves at 2am.

Today I work AI-first and build agentic AI applications for clients. Claude, GitHub Copilot, GPT and Gemini are part of my daily toolchain — not as autocomplete, but as collaborators for architecture, refactoring and shipping. The eleven years are what make that work: knowing what to ask for, and knowing when the answer is wrong.

I lead engineering teams building Flutter and web applications for clients. Earlier in my career, I worked on delivering apps for global brands such as OMRON and Emerson. Outside of client work I run product and the software platform at SPDrone.tech, my family’s drone company. I also write on Medium, answer questions on Stack Overflow and ship personal products.

  • NowTech Lead · Netsmartz IT Solutions
  • AlsoProduct & platform · SPDrone.tech, the family drone company
  • BasedMohali, India — working globally
  • FocusAgentic AI applications · Flutter architecture
  • SideIndie apps · technical writing

Star Performer of the Quarter

Netsmartz IT Solutions

Gold Medal — Aero-modelling

NCC aerobatics competition, North India

02 How I build

AI-first, not AI-only.

I treat frontier models as engineering collaborators — scaffolding, refactoring and exploring in parallel — while the architecture, the review and the accountability stay mine. Eleven years of shipping is what makes the difference between fast and reckless.

Claude Code

Agentic development in the terminal — multi-file refactors, test generation and codebase-wide reasoning on real production repositories.

Daily driver

GitHub Copilot

In-editor pair programming — completions, chat and agent mode. With Claude, the foundation of my AI-first workflow on client delivery.

Client delivery

GPT

Rapid problem framing, API design exploration and turning half-formed product ideas into specifications worth building from.

Ideation & specs

Gemini

In-product intelligence. Gemini AI Studio and the Gemini API power the assistants inside DayFlow, Antigravity Chat and my AI Recipe Assistant.

Shipped in production

Antigravity

Agent-driven IDE workflows — and the platform behind my Antigravity Chat build on Gemini 3.1 Pro.

Agentic IDE

Firebase Studio

AI-assisted full-stack prototyping — going from prompt to a deployed, authenticated app while the idea is still warm.

Prototyping

On-device ML

Google ML Kit and embedded models — vision, text and inference that keep working when the network doesn’t.

Edge AI
  1. 01

    Frame

    Define the problem and constraints precisely — before a single token is generated.

  2. 02

    Pair

    Run agents on scaffolding, migrations and the tedious 80% in parallel.

  3. 03

    Review

    Read every line. Architecture, security and edge cases are not delegated.

  4. 04

    Ship

    Test, instrument, release — and own what happens next.

03 Experience

Eleven years, four teams, one throughline.

From Android developer to Tech Lead. Select any role to expand it.

  1. Leading delivery for enterprise clients, building agentic AI web and Flutter applications, mentoring engineers, and driving technical decisions from proposal through release, with an AI-first workflow built on Claude and GitHub Copilot.

    • AI-first delivery
    • Agentic AI
    • GitHub Copilot
    • Flutter
    • Dart
    • Kotlin
    • BLoC
    • Firebase
    • Team leadership

    Star Performer of the Quarter

  2. Analysed and implemented technology solutions across client engagements — evaluating approaches, prototyping, and translating business requirements into working architecture.

    • Solution analysis
    • Architecture
    • Android
    • Prototyping
  3. Led the Android development team — owning project oversight, code quality, release cadence and day-to-day technical direction.

    • Android
    • Kotlin
    • Team leadership
    • Code review
  4. Where it started. Over four years building and maintaining Android applications across a wide range of client domains — the foundation for everything since.

    • Android SDK
    • Java
    • BLE / NFC
    • Material Design
    • MVVM

04 Selected work

Things I’ve built and shipped.

Featured personal project — live

Should I Buy This Car?

What the car actually costs, measured against what you actually earn.

A car affordability calculator for India that refuses to stop at the EMI. It builds the on‑road price up from its parts, models ten years of ownership — fuel or charging, insurance against a falling IDV, upkeep that rises with age, resale credited back — and then writes the verdict in plain English rather than leaving you a table to interpret.

  • A written verdict — graded against take‑home pay during the loan years, not a lifetime average, with the 20 / 4 / 10 rule checked line by line
  • 103 Indian models — petrol, diesel, hybrid and EV, with real‑world mileage factored in against the ARAI figure
  • EV vs petrol — same loan, same distance, same horizon, with the break‑even year on a cumulative‑spend chart
  • The alternatives — buy new, buy three years old, lease, or don’t buy and invest the difference
  • Honest modelling — income tax recalculated on each year’s pay, running costs inflated year by year, and a flat‑salary stress test
  • Vanilla JS
  • ES modules
  • Custom SVG charts
  • Financial modelling
  • Vercel

05 Toolkit

What I reach for.

Eleven years of accumulated preferences, grouped by where they live.

✦ AI-First Development

  • Claude Code
  • GitHub Copilot
  • GPT
  • Gemini
  • Antigravity
  • Firebase Studio
  • Agentic AI
  • Agentic workflows
  • Prompt engineering
  • AI code review

📱 Mobile

  • Flutter
  • Dart
  • Android SDK
  • Kotlin
  • iOS
  • BLE
  • NFC
  • IPC
  • Material Design

🧠 AI & ML

  • Google ML Kit
  • Gemini AI Studio
  • On-device ML
  • Generative AI
  • LLM integration

🏗️ Architecture

  • BLoC
  • MVVM
  • Design patterns
  • Clean architecture
  • Root cause analysis

☁️ Web & Cloud

  • Next.js
  • Tailwind CSS
  • Firebase
  • Firebase Studio
  • Google Cloud
  • Vercel

🛠️ Tooling

  • Git / GitHub
  • Azure DevOps
  • Jira
  • Play Console
  • CI/CD

🤝 Leadership

  • Team leadership
  • Mentoring
  • Technical strategy
  • Client delivery
  • Code review

07 Contact

Let’s build
something good.

Consulting, a leadership role, a tricky Flutter problem, or a drone idea — I read everything that lands in my inbox.

Malwinder Singh Tech Lead · Mohali, India (IST)
Location Mohali, Punjab — India (IST, UTC+5:30)

Goes straight to my inbox — I usually reply within a day.