I'm Bushra Khandoker — a Computer Science student at the University of Greenwich (UK, Year 3), building at the intersection of systems engineering, AI/ML, and product development.
I don't just write code. I architect production systems, study machine learning from first principles, and solve real problems with measurable impact. My work spans from bKash-integrated e-commerce platforms in Bangladesh to interactive algorithm visualizers — each project grounded in genuine engineering depth.
Currently preparing for AI/ML engineering and software engineering roles at top-tier UK tech companies, while simultaneously working through a rigorous self-directed curriculum spanning algorithms, probability theory, linear algebra, and deep learning.
Bushal — Full-Stack E-Commerce Platform · Bangladesh
Heritage-quality, curated products delivered across Bangladesh with secure bKash payments.
A production-grade e-commerce platform built as a serious engineering portfolio piece — not a tutorial clone.
Next.js 14 (App Router) · Supabase (PostgreSQL + Auth + RLS)
TypeScript · Tailwind CSS · bKash PGW Integration
Midnight Market Design System · Vercel Edge Deployment
Engineering highlights:
- Atomic stock management via PostgreSQL RPC (prevents race conditions at the DB level)
- Row-Level Security policies with SSR-safe
getUser()auth pattern - bKash Payment Gateway callback handler with idempotency guard
- Custom middleware using Supabase SSR pattern for protected routes
- "Midnight Market" design system: deep navy + warm gold + Cormorant Garamond typography
PhyCode — Interactive Algorithm Learning Platform
Making algorithms feel less abstract and more alive.
Next.js · TypeScript · Tailwind CSS · Recoil · Firebase Auth
What makes it different: Visual, step-by-step algorithm execution with user accounts, progress tracking, and a responsive design built to teach — not just display. Inspired by the insight that deep understanding requires seeing a system move, not just reading its description.
StudyBuddy — Study Session Tracker with AI Features
Precision study tracking with a Pyodide/SymPy CAS calculator, multi-theme system, and Firebase sync.
React · Firebase Auth + Firestore · Pyodide · SymPy · IndexedDB
Engineering highlights:
- Computer algebra system (CAS) embedded via Pyodide with fast-first fallback and IndexedDB caching
- Multi-theme engine (5 themes: Pookie, Aesthetic, Steller, Notebook, Graph Paper) with canvas animations and custom cursors
- Magic link auth flow (OTPAuth) with production-safe Vercel deployment
- Firebase Firestore schema structured to eliminate composite index requirements
Languages
| Language | Proficiency | Primary Use |
|---|---|---|
| C++ | Advanced | DSA, competitive programming, STL mastery |
| Python | Intermediate–Advanced | ML pipelines, scripting, SymPy/Pyodide |
| TypeScript | Strong | Full-stack web, type-safe APIs |
| JavaScript | Strong | Frontend, Node.js, Firebase SDK |
| SQL (PostgreSQL) | Intermediate | Supabase, RLS, stored procedures, RPCs |
Frontend
Next.js 14 (App Router) · React · TypeScript · Tailwind CSS
Recoil · HTML5 · CSS3 · Framer Motion · Shadcn/ui
Backend & Infrastructure
Supabase (PostgreSQL + Auth + Storage + Edge Functions)
Firebase (Auth, Firestore, Realtime DB)
Node.js · REST APIs · Vercel Edge Deployment
bKash Payment Gateway (PGW) · Row-Level Security (RLS)
Algorithms & Data Structures
Core structures
Arrays · Strings · Linked Lists · Stacks · Queues
Trees · BSTs · Heaps · Graphs · Hash Maps · Tries · Disjoint Sets
Techniques
Dynamic Programming · Greedy · Divide & Conquer · Backtracking
Binary Search · Two Pointers · Sliding Window · Graph Traversals (BFS/DFS)
Topological Sort · Segment Trees · Union-Find
AI / ML Foundations
Linear Algebra (MIT 18.06) → matrix decompositions, eigenspaces, SVD
Probability Theory (Harvard Stat110) → conditional expectation, LOTUS, MGFs
Statistics (MIT 18.650) → MLE, hypothesis testing, regression
ML Theory (Stanford CS229) → gradient descent, SVM, neural nets
Deep Learning (Goodfellow et al.) → backprop, CNNs, regularisation
Pattern Recognition (Bishop PRML) → probabilistic graphical models
Maintaining a structured Obsidian knowledge vault connecting linear algebra foundations → ML theory → implementation — built to support dissertation work and internship positioning.