Raipur, India
2026
Anmol Singh TutejaApps, websites, desktop tools, interfaces, and the automation nobody wants to touch. Built to survive a bad network and still be standing six months later.
Mobile apps, Web apps, Desktop tools, Interface design, Internal tools, Automation, Play Store delivery, Performance work
Mobile apps, Web apps, Desktop tools, Interface design, Internal tools, Automation, Play Store delivery, Performance work
01 / What I do
The tools change every two years. The job does not. Somebody has a problem that costs them time or money and needs software that makes it stop.
You know what is broken. I turn that into flows, screens and something you can click, before a line of production code exists. Cheaper to change a Figma file than a shipped app.
Android and iOS from one codebase. Built for real Indian conditions: patchy networks, cheap devices, and users who close the app halfway through a task and expect to come back to it.
Marketing sites, customer portals, dashboards. Fast on a 3G connection in a small town, not just on your laptop with fibre.
The unglamorous internal thing your team actually runs the business on. Nobody wants to build it. I will.
The job somebody does by hand every Monday morning. Hand it to a script and take the morning back.
02 / Selected work
5 case studies
Enterprise field operations2024 to now
The platform more than 100 employees open before anything else, with location tracking that keeps working after the app is killed.
Consumer ISP app2024
Rebuilt a consumer ISP app from nothing, before Google removed the old one from the Play Store.
Desktop product2026
A desktop product that keeps WhatsApp automation running without anyone watching it, licensed per machine and sold to paying customers.
Freelance, staff performance2024
Turned staff productivity from an argument at the end of the month into a number both sides can see.
Personal product2023
A dictionary that still works when the network does not.
03 / The workbench
These are not client work and I am not presenting them as results. They are what I do when I want to find out whether something is possible. Most of them are live and you can open them right now.
Live3D mockup tool2026
Drop a screenshot on a phone, a laptop or a watch and export the scene as an image or a video.
The whole scene is WebGL. Rendering, animation and encoding all happen in the browser, so a video is composited and muxed on the visitor machine rather than on a server. Real device models where the shape matters, procedural geometry where it does not. Sign in is the only thing that leaves the browser.
LiveVideo and audio editor2026
A video editor for people who make videos, with no upload and no account.
Every cut, every render and every export runs on the machine it is opened on. There is no server behind it, which means nothing to pay for, nothing to breach, and a file that never leaves the room it was recorded in. Free to use.
LiveImage privacy tool2026
Takes the location, the device and the invisible watermarks out of a photograph before you post it.
A photo carries more than the picture. Where it was taken, which camera took it, sometimes a watermark you cannot see. This strips all of it, and the deep clean redraws the image so nothing survives in the file. The photo is never uploaded, because the work happens in the browser and it never leaves the machine.
LiveDesign reference2026
Colour theory you can turn the dial on instead of reading about.
A twelve segment wheel you pick a hue on, six harmonies worked out from whatever you picked, colour psychology, curated palettes, and the colour models shown with live visuals rather than diagrams. The picker runs RGB and HSB sliders that move together, so you can watch what each model does to the same colour.
BuiltAnimation editor2026
A Lottie editor where the animation engine underneath it is mine.
The shape renderer, the keyframe interpolation and the easing are all written in the project rather than pulled in. The only animation dependency is the official player, used once in the export dialog to prove the file it just wrote actually plays.
04 / Receipts
0%
faster processing
Image to location dropped from 26 seconds to 1 or 2 seconds, after I replaced a slow third-party package with a tool I built.
₹0/yr
licence cost removed
Persistent location tracking that survives app termination, usually a paid SDK. I built it in house instead.
0+
employees, every day
Field operations platform I own end to end. Attendance, tickets, collections, commute tracking.
0%
faster delivery
After moving the core codebase to Clean Architecture and MVVM, features started shipping measurably quicker.
05 / How I work
It breaks when nobody said out loud what was being built. This is how I avoid that.
One conversation. I ask what goes wrong today and what it costs you. If software is not the answer, I will tell you that instead of selling you an app.
We agree on the smallest version worth putting in front of real users, and we write down what it will not do. Nobody is guessing later.
You get builds while they are being built, not a reveal at the end. Changes are cheap early and painful late.
Release, watch what real users do to it, fix what they find. An app that ships and then rots helped nobody.
06 / Writing and video
I write about the parts of Flutter that are badly documented, the ones I had to work out the hard way. Sixteen articles so far, and the most read one has crossed 900 reads.
07 / About
2024 to now
Elxer Communications
Lead Mobile App Developer
2023
Xhitiz IT Solutions
Senior Flutter Developer
2022 to 2023
Datalytics
Mobile App Developer
I am Anmol. I live in Raipur. I taught myself Python in 2015, from tutorials and documentation, with nobody setting the assignments. My first public code went up in January 2017, in Java, PHP, C and C++.
I have been building mobile software professionally since 2022.
Right now I am the only mobile engineer at Elxer Communications, which means I own two Flutter products completely: an enterprise field operations platform used daily by more than 100 employees, and a consumer ISP app I rebuilt from nothing after the previous one nearly got pulled off the Play Store.
Before that I built e-commerce, attendance systems and an Android TV app that streamed live camera feeds. Alongside the job I take on work that looks interesting, and I write about the messy parts of the craft.
Mobile is what I am paid for most, so it is what people assume I do. It is not the whole of it. Count my own repositories and TypeScript is now slightly ahead of Dart, with Python, Kotlin, Swift and a little Rust behind them. I pick the thing the problem needs.
I do not have a favourite framework. I have a preference for problems that are worth solving, and a low tolerance for software that looks finished but falls apart the first time a real user touches it.
On the record
What I reach for
Also shipped
Sanghathit
Social platform with hashtag discovery, multi-image and video posts, and real-time audio and video calling through Agora. Live on Google Play.
RJU e-commerce
Cart, wishlist, payment gateway and order management, built pixel-accurate from Figma with BLoC.
Workforce attendance
Live location capture, photo verification and device ID binding, used daily by field staff.
Android TV streaming
Auto-start on boot, RTSP live camera streaming, and server-driven UI so content changed without a release.
08 / Contact
If I can fix it, I will tell you how and roughly what it takes. If I cannot, I will say so and point you at someone better placed. Either way you get a straight answer, not a pitch.