I build production web apps, end to end.
Full-stack engineer with 3+ years shipping React, Next.js, Node.js, NestJS and FastAPI applications — from AI tutoring platforms and headless CMSs to SaaS products and admin systems.
Dhaka, BangladeshOpen to new opportunities
- TypeScript
- React
- Next.js
- Node.js
- NestJS
- Python
- FastAPI
- MongoDB
- PostgreSQL
- Redis
- Docker
- Tailwind CSS
- Gemini APIs
- RAG
Engineer across the stack,
focused on production.
I build and maintain production web applications across frontend and backend — REST APIs, authentication and RBAC, database-driven features, caching, performance work and production debugging.
I've built AI-powered products, headless CMS platforms, SaaS applications and administrative systems. I'm comfortable dropping into unfamiliar codebases, owning well-scoped features, reviewing code and working closely with product and engineering teams.
- 3+
Years in production
Building and maintaining live web applications for real users.
- 90%
Less revalidation compute
Redesigned ISR and caching in a production Next.js app.
- $5
Monthly infra, down from $200+
Same app, same traffic — after the caching overhaul.
Where I've
shipped.
Innospace Infotech Ltd.
Software Engineer (Full Stack)Sep 2025 – PresentDhaka, Bangladesh- Develop and maintain production web applications across frontend and backend with React, Next.js, TypeScript, Node.js and NestJS.
- Build REST APIs, authentication flows, authorization, RBAC, validation and database-backed features.
- Contribute to frontend architecture, reusable components, SSR, caching and performance in Next.js apps.
- Work with MongoDB data models, schema design and query optimization; contribute to Docker-based deployment workflows.
Themefisher
Full-Stack DeveloperMay 2023 – Sep 2025Shyamoli, Mohammadpur, Dhaka- Progressed from intern-level frontend work to full-stack development on production SaaS and web applications.
- Primary engineer on Sitepins (headless CMS) and contributor to UIHut (design SaaS) across frontend, backend, auth and admin.
- Built production features with React, Next.js, Node.js, TypeScript, MongoDB and REST APIs.
- Translated requirements into maintainable software alongside engineering and product teams.
Things I've
built.
AI products, a headless CMS and SaaS platforms, all running in production. Here's what I built and the stack behind each one.
More on GitHub- brritto.com(opens brritto.com)AI-powered learning platform
Brritto AI Tutor
An AI tutoring platform with streamed chat, retrieval-based answers, and subscription and admin workflows.
- AI chat streaming over Server-Sent Events for incremental responses.
- Access/refresh tokens in HttpOnly cookies, protected routes and RBAC.
- MongoDB Atlas vector search for RAG, plus Redis, Langfuse and S3/MinIO.
- Next.js
- FastAPI
- MongoDB Atlas
- Redis
- Gemini
- SSE
- sitepins.com(opens sitepins.com)Headless CMS · Multi-site platform
Sitepins
A headless CMS for managing content across many websites, where I was a primary engineer.
- Full-stack features, auth and an administrative dashboard.
- Role-based access control and APIs for multi-site content workflows.
- Architecture, production debugging and reusable components.
- Next.js
- Node.js
- MongoDB
- REST APIs
- RBAC
- uihut.com(opens uihut.com)Design SaaS platform
UIHut
A design-resource SaaS where I contributed production features across frontend and backend services.
- Frontend experiences and backend services, including authentication.
- Reusable UI components while keeping existing features stable.
- React
- Next.js
- Node.js
- MongoDB
- bapusWeb application
BAPUS
Production application work covering feature implementation, bug fixing, maintenance and backend service integration.
- Features and fixes in React/Next.js against backend APIs.
- React
- Next.js
- REST APIs
The toolkit I
reach for.
TypeScript across the stack, Python where AI work needs it, and the infrastructure to ship and run both. I pick up new tools quickly when a project calls for them.
Languages
- TypeScript
- JavaScript
- Python
- SQL
- HTML
- CSS
Frontend
- React
- Next.js
- Server Components
- SSR / SSG / ISR
- Tailwind CSS
Backend
- Node.js
- NestJS
- Express
- FastAPI
- REST & API design
Data & Infra
- MongoDB
- PostgreSQL
- Redis
- Docker
- S3 / MinIO
- Vercel
- CI/CD
Architecture
- Auth & RBAC
- Caching
- Background jobs
- API security
- Headless CMS
AI Engineering
- Gemini APIs
- Streaming responses
- RAG
- Vector search
- Langfuse
- B.Sc. in Computer Science & EngineeringEuropean University Bangladesh
- Diploma in Computer Science & TechnologyTangail Polytechnic Institute
What you get
on day one.
An engineer who owns features end to end, cares about what they cost to run, and keeps shipping after launch day.
I own features end to end
Schema, API, auth, UI and deployment. You don't need a separate frontend and backend hire to get a feature into production.
Primary engineer on Sitepins, a multi-site headless CMS.
I treat cost and speed as features
I look for where the money and milliseconds go, then fix the architecture rather than adding more servers.
Cut a Next.js app's infra bill from $200+ to $5 a month.
I ship AI that holds up in production
Streaming, retrieval, observability and access control. Not a demo wrapped around an API key, but a product people pay for.
Built Brritto AI Tutor with SSE streaming and vector-search RAG.
I get productive in your codebase fast
I read before I write, keep existing features stable, and leave code easier for the next person to change.
Grew from intern to full-stack engineer on live SaaS products.
Notes from
the work.
Write-ups on things I've built and debugged in production — streaming AI responses, retrieval, auth, caching and deployment.
All posts- 9 min read
B-Tree কীভাবে কাজ করে: Binary Tree থেকে Database Index পর্যন্ত
B-Tree কেন একটা node-এ অনেক key রাখে, n key কেন n+1 child বানায়, tree কীভাবে নিজে balanced থাকে, আর database index-এর জন্য B-Tree ও B+ Tree কেন সবচেয়ে উপযুক্ত।
- 15 min read
PostgreSQL Internals, Part 5: Transaction, Isolation আর WAL
Transaction আর ACID কী, isolation level অনুযায়ী snapshot কীভাবে বদলায়, আর WAL ও checkpoint মিলে PostgreSQL কীভাবে crash-এর পরেও committed data বাঁচিয়ে রাখে, সেটা এখানে বুঝব।
- 13 min read
PostgreSQL Internals, Part 4: Architecture আর Query Lifecycle
একটা connection খোলা থেকে result পাওয়া পর্যন্ত PostgreSQL-এর ভেতরে কোন process কী করে, shared buffers আর dirty page কী, আর একটা query কোন কোন ধাপ পেরিয়ে চলে, সেটা এই পর্বে বুঝব।