Skip to content
Kamran Khan — home

Kamran Khan

Full-Stack Software Engineer

Associate Software Engineer at GDA

I build enterprise web applications end to end — ASP.NET Core APIs, React front ends and the PostgreSQL / SQL Server schemas underneath — and integrate AI into products where it earns its place.

View workGitHub (opens in a new tab)
~/kamran — zsh
whoami
Kamran Khan — full-stack software engineer
cat ./current-role
Associate Software Engineer @ Galiyat Development Authority
ls ./stack/core
C#ASP.NET CoreEntity Framework CoreREST API designReactTypeScriptPostgreSQLSQL Server
based in
Swat, Pakistan
experience
3+ years
education
BS Software Engineering

01 / experience

Where I work

Enterprise software for a government authority: back-end services, REST APIs, data models and the front end on top of them.

  1. Present

    Full-time

    Abbottabad, Pakistan

    Associate Software Engineerat Galiyat Development Authority

    Building and maintaining the authority's enterprise ERP: back-end services, REST APIs, database work and the React front end.

    • Develop ERP modules for Asset Management, Inventory, Finance, HR, Property and Procurement, including their business workflows.
    • Design and build REST APIs and back-end services in ASP.NET Core with Entity Framework Core.
    • Apply Clean Architecture and a modular design across the codebase.
    • Implement role-based access control (RBAC).
    • Model and maintain relational schemas on PostgreSQL and SQL Server; containerise services with Docker.
    • C#
    • ASP.NET Core
    • EF Core
    • React
    • TypeScript
    • PostgreSQL
    • SQL Server
    • Docker
    Read the case study

02 / work

Selected work

Work on an enterprise ERP, my final-year project, and personal projects. Each one has a case study: what it does, how it is put together and why.
ProfessionalOngoing
Associate Software Engineer @ GDA

Enterprise ERP Platform

Multi-module ERP for a government development authority.

An ERP platform covering asset management, inventory, finance, HR, property and procurement, with business workflows and role-based access control.

  1. Client

    • React
    • TypeScript

  2. API

    • ASP.NET Core
    • REST
    • RBAC

  3. Domain

    • Clean Architecture
    • Module per domain
    • EF Core

  4. Data

    • PostgreSQL
    • SQL Server
  • C#
  • ASP.NET Core
  • EF Core
  • React
  • TypeScript
  • PostgreSQL
  • SQL Server
  • Docker
Case study: Enterprise ERP Platform

Internal system — source and screenshots are the employer's property.

Final-year projectCompleted
BS Software Engineering, final-year project

CyTechPro — Hybrid Intrusion Detection System

Full-stack web app that wraps an ML pipeline for network intrusion detection.

My final-year project: an ensemble of ML models served through a FastAPI back end, with live packet capture, traffic-file analysis, threat alerts and a web dashboard.

  1. Capture

    • TShark
    • Traffic upload

  2. API

    • FastAPI
    • Auth

  3. Models

    • Random Forest
    • XGBoost
    • CatBoost
    • LightGBM

  4. Client

    • Dashboard
    • Alerts
    • Export
  • Python
  • FastAPI
  • TShark
  • scikit-learn
  • XGBoost
  • CatBoost
  • LightGBM
Case study: CyTechPro — Hybrid Intrusion Detection SystemLive: CyTechPro — Hybrid Intrusion Detection System (opens in a new tab)Code: CyTechPro — Hybrid Intrusion Detection System (opens in a new tab)

Full source is private; the linked repo is a public showcase with screenshots.

Personal projectIn development

ElevareCV — AI Résumé & Career Platform

Résumé builder with job matching, tailoring and grounded AI suggestions.

A career-management app: master career profile, résumé versions, import, PDF/DOCX/TXT export, job matching and tailoring, cover letters and an application tracker. This repo is the React web client; the ASP.NET Core API it talks to is specified in docs/API.md and typed in TypeScript contracts.

  1. Client

    • React 19
    • TanStack Query
    • Zustand
    • PDF / DOCX / TXT renderers

  2. API (specified)

    • ASP.NET Core
    • docs/API.md
    • Typed contracts

  3. AI

    • Server-side only
    • Consent + grounding check
  • React 19
  • TypeScript
  • Vite
  • TanStack Query
  • Zustand
  • Tailwind
  • Vitest
  • MSW
  • Playwright
Case study: ElevareCV — AI Résumé & Career PlatformCode: ElevareCV — AI Résumé & Career Platform (opens in a new tab)

Also built

03 / stack

Tools I build with

Grouped by depth, not by count. Core is what I'd lead a feature in; working is what I've used in real projects; the last group is where AI shows up in my apps.

core/

08 items

What I build with every day and would lead a feature in.

  • C#
  • ASP.NET Core
  • Entity Framework Core
  • REST API design
  • React
  • TypeScript
  • PostgreSQL
  • SQL Server

working/

13 items

Used in real projects; comfortable picking back up.

  • JavaScript
  • Next.js
  • Node.js / Express
  • Python
  • FastAPI
  • Prisma
  • MongoDB
  • Docker
  • Tailwind CSS
  • TanStack Query
  • Vitest
  • Playwright
  • Git

applied/ AI in products

04 items

Putting trained models and LLM APIs behind services in the apps I build.

  • LLM API integration (server-side)
  • Serving ML models behind FastAPI
  • scikit-learn
  • XGBoost / CatBoost / LightGBM

practices

  • Clean Architecture
  • Modular design
  • Role-based access control
  • Relational data modelling
  • Typed API contracts
  • Automated testing

04 / approach

How I build software

A few working rules, each tied to a project where it actually applies.

  1. Boundaries first

    I start with the modules, the data and who is allowed to do what. Clean Architecture and explicit RBAC make a system easy to extend one domain at a time.

    seen in: Enterprise ERP

  2. Typed contracts end to end

    The API shape is written down before it is built and mirrored in TypeScript contracts, so the client is coded against an agreed shape instead of guesses.

    seen in: ElevareCV

  3. Test where it breaks

    Tests go where bugs are expensive: HTTP clients, exports, autosave and conflict handling — unit and integration first, end-to-end for the critical paths.

    seen in: ElevareCV

  4. AI behind a service boundary

    Models and LLM calls sit server-side behind an API with clear inputs and outputs. Where AI writes text for a user, its output is checked against the user's own data before it is shown.

    seen in: CyTechPro IDS · ElevareCV

05 / about

How I got here

I'm a full-stack software engineer. My day job is enterprise software: I work on an ERP platform covering asset management, inventory, finance, HR, property and procurement — the kind of system where data integrity, permissions and long-lived maintainability matter more than anything flashy.

My strongest stack is C# / ASP.NET Core on the back end and React with TypeScript on the front end, backed by relational databases. I also build AI-powered applications — wiring ML models and LLM APIs behind well-defined services — as a full-stack engineer, not a research scientist.

I learned to program the long way: C, C++ and Java (OOP, data structures) before moving to the web and .NET. That foundation is why I care about clear boundaries, typed contracts and code the next person can change safely.

06 / contact

Get in touch

The best way to reach me is email — or use the form, which lands in the same inbox.

Direct channels

POST /contact

2–100 chars

used only to reply

0 / 5000

all fields required