Pinty Jangid · engineering

Services

What I build, and how we would work.

Six things I do well. Everything below has shipped to real users on a real deadline — the capability list is short on purpose.

01

Product engineering

A complete web application, front to back. I own the data model, the API, the interface and the way they fit together, so there is one person accountable when something does not line up.

  • Greenfield products from an idea and a whiteboard
  • Rebuilds of software that has outgrown its first version
  • Authentication, roles, billing, notifications, documents, reporting
  • Multi-language and multi-region support built in from the start
AngularReactSpring BootFastAPIPostgreSQL

02

Mobile apps

One Flutter codebase producing an iOS and an Android app that share their logic with your web product. I take it all the way through review and into the stores.

  • Native-feeling apps for both platforms from a single codebase
  • Signing certificates, provisioning, store listings and review responses
  • Staged rollouts, crash reporting, over-the-air config
  • Offline behaviour and push notifications where they earn their place
FlutterApp StorePlay StoreFastlane

03

AI features that behave

The interesting part of an AI feature is not the prompt. It is what happens when the model is wrong — and whether your users can tell. I build the parts that make it safe to ship.

  • Summaries, trend analysis, risk scoring and classification over your own data
  • Retrieval over documents, with answers traced back to their source
  • Citation checks and guardrails that reject unsupported output in code, not by hope
  • Evaluation sets, so a prompt change is measured rather than argued about
LLM APIsRAGPythonEvaluation

04

Developer tooling & codebase intelligence

For teams sitting on a large codebase nobody fully understands. I build the tools that make its shape visible, and the gates that stop it getting worse.

  • Architecture maps and dependency graphs read from the source itself
  • Hotspot analysis from git history — where the risk actually concentrates
  • CI checks that fail a pull request for the problem it added, not the ones it inherited
  • Internal CLIs and agent-facing tooling for your own developers
TypeScriptStatic analysisGitHub ActionsMCP

05

Interface & design systems

An interface is a promise about how much you care. I build the visual system underneath it, so quality survives the tenth screen and the third developer.

  • Design tokens for colour, type, spacing and elevation
  • A reusable component library your team can build with
  • Reskins and modernisation of existing products, rolled out in phases
  • Keyboard access, contrast and screen-reader behaviour treated as requirements
Design tokensComponentsAccessibilityFigma to code

06

Cloud, deployment & ongoing care

Software that is not deployed is not finished. I set up the path from a commit to production and stay reachable once real people depend on it.

  • Containerised deployments, staging and production environments
  • Build and release pipelines, database migrations, backups
  • Logging, monitoring and alerting that names the actual problem
  • Retained support and incident response after launch
DockerAWSCI/CDMonitoring

Engagements

Three ways to start.

Every project is quoted against a written scope, so you know the number before any work begins. No hourly meter running in the background.

Option A

Discovery sprint

One to two weeks, fixed price. For when the idea is clear but the shape is not.

  • Requirements pinned down in writing
  • Architecture and technology decisions, with the reasoning
  • A clickable prototype of the core flow
  • A costed, phased build plan

You keep everything, including the plan — even if you build it elsewhere.

Option B

Build engagement

Typically six to sixteen weeks. The main way I work: a defined product, built and launched.

  • Fixed scope, milestone billing, agreed launch date
  • Staging environment live from week one
  • Weekly written progress and a working build to click
  • Production deployment, documentation and handover included

Scope changes are re-quoted openly, never absorbed silently into the timeline.

Option C

Ongoing partnership

A monthly retainer. For products already live that need a steady hand rather than a rebuild.

  • A reserved block of days each month
  • New features, improvements and technical debt paid down
  • Store releases, dependency and security updates
  • Incident response with an agreed response window

Month to month. If it stops being worth it, you stop.

A good fit

When this works well

  • You have a real problem and a decision-maker who can answer questions in a day.
  • You would rather have one senior person accountable than a rotating team of five.
  • The product matters enough to deploy, monitor and maintain properly.
  • You want to own the code, the accounts and the pipeline at the end.

Not a good fit

When to hire someone else

  • You need twenty developers next month — that is an agency, not an individual.
  • The scope is "everything Amazon does" on a four-week budget.
  • The work is a pixel-for-pixel clone of a competitor with no product thinking involved.
  • Nobody on your side can be reached for decisions. Silence stalls builds faster than bugs.

Questions

The things people ask before the first call.

What does a project cost?

It depends entirely on scope, so I quote per project rather than publishing a rate card. After a short call you get a written scope with a fixed number attached. Small, well-defined pieces of work are quoted as a single figure; larger builds are split into milestones you approve and pay as they complete.

How long until something is live?

A discovery sprint runs one to two weeks. A first usable version of a focused product is usually six to ten weeks; larger platforms with mobile apps and multiple user types run longer. You will have a staging URL to click within the first week either way.

Can you work with our existing team and codebase?

Yes, and often that is the better arrangement. I join your repository, your board and your review process, and work the way your team already works rather than importing my own.

Who owns the code?

You do — fully, from the first commit. It lives in your repository under your accounts, and the handover includes documentation written for whoever maintains it next. No proprietary framework you can only run through me.

What happens after launch?

Every build includes a warranty window for anything that turns out to be broken. Beyond that, you can take it in-house with documentation, or keep me on a monthly retainer for features, releases and incidents.

Do you sign NDAs and work with regulated data?

Yes to NDAs, before any detail is discussed. I have built healthcare products where confidentiality and careful data handling were requirements rather than preferences, and I am happy to work inside your compliance constraints and review process.

Which time zones do you work across?

I work with clients across time zones and keep a reliable overlap window for calls, with written updates covering the rest. Deadlines and response windows are agreed in writing before we start.

Next step

Tell me what you are trying to build.

A half-hour call, then a written scope with a real number in it. No obligation attached to either.

Get a scope