SP
Software engineer · AI systems builder

Md. Salman HossanProttoy.

Software Engineer+AI Systems Builder

I turn messy product problems into reliable software — then make AI useful in the real world.

Jyväskylä, FinlandOpen to software & AI opportunities
Let's talkResumeExplore the work
Systems tracelive / 08.26
Md. Salman Hossan Prottoy

Builder profile

Md. Salman Hossan Prottoy

Human in the loop · thesis underway

Ships with

Care + clarity

Reach me

salman.prottoy@gmail.com

About

Useful software, thoughtfully made.

A little context on how I work and what I bring to a team.

I am a software engineer and Master's student in Artificial Intelligence at the University of Jyväskylä, Finland, currently starting my Master's thesis in multi-objective optimization. I build full-stack products with TypeScript, React, Go, Python, and AWS, with a particular interest in making language-model systems reliable enough to use beyond a demo.

Most recently at Brand Cloud Inc., I worked across production TypeScript/React systems, zero-downtime PostgreSQL migrations, CI/CD, and a RAG application used by 500+ people. Earlier at W3 Engineers, I helped move backend services from Django to Go and connected data-heavy systems across a multi-tenant travel platform.

Build for the happy path. Design for everything that happens after it.

Selected signals

500+

RAG users reached

2,000+

documents / hour at peak

40%

travel platform endpoints

35%

monitoring frontend

Experience

Where I have shipped.

Production systems, migrations, and product work across Japan and Bangladesh.

Software engineering · full stack · infrastructure · applied AI

Software Engineer

Brand Cloud Inc. · Tokyo, Japan (Remote)
September 2023 — June 2026
  • Migrated CMS content storage to PostgreSQL on Amazon RDS with zero downtime for end users, integrating backend services across the platform.
  • Grew a RAG-based application to 500+ users and 2,000+ documents processed per hour at peak, combining vector search, caching, and containerized Docker infrastructure to let users query multiple LLMs over uploaded documents.
  • Designed and maintained automated GitHub CI/CD pipelines that validated changes before merge, speeding up pull-request review while following DevSecOps practices.
  • Mentored 2–3 junior engineers and established team-wide coding standards, improving sprint delivery consistency.
  • Shipped full-stack, business-critical features on a TypeScript/React (Next.js) stack deployed and maintained on Vercel.

Junior Software Engineer

W3 Engineers Limited · Dhaka, Bangladesh
November 2021 — August 2023
  • Contributed client-server integrations via REST and GraphQL APIs in Go and Node.js, covering roughly 40% of the platform's endpoints for a 30+ domain multi-tenant travel platform.
  • Contributed to the backend migration from Django to Go, helping significantly reduce response latency across the platform's core services.
  • Built data pipelines to unify property data from multiple external providers into a single schema, using parallel processing and Amazon SQS asynchronous messaging on AWS.
  • Built reusable, production Vue.js UI components powering 35% of the frontend of an enterprise network-monitoring application, integrating backend APIs to improve the end-user experience.
Selected work

Work with a signal.

A cross-section of production engineering, applied AI, distributed data, and published research.

500+

users

Production AI · Brand Cloud

RAG application

A document intelligence workflow where users query multiple LLMs over uploaded content. The system combines retrieval, caching, and Docker-based services into a product that can handle real usage rather than just a notebook demo.

RAGLLMsVector searchDocker
Private case study

0

downtime

Platform reliability · Brand Cloud

CMS data migration

Moved CMS content storage to PostgreSQL on Amazon RDS while keeping the product available. Backend services across the platform were integrated around the new source of truth without a disruptive cutover.

PostgreSQLAmazon RDSTypeScriptZero downtime
Private case study

30+

domains

Distributed systems · W3 Engineers

Travel data platform

Unified property data from multiple external providers into one schema, using parallel processing and Amazon SQS asynchronous messaging across a multi-tenant travel platform.

GoNode.jsRESTGraphQLAWS SQS
Private case study

2022

published

Published research · IC4IR 2021

Bangla fake-news detection

Explored imbalance handling and model stacking for fake-news classification in Bangla, contributing to research published in Springer’s Lecture Notes in Networks and Systems.

PythonNLPTensorFlowModel stacking
Publication
Education

Always still learning.

The academic foundation behind my curiosity for intelligent systems, including my current thesis work.

2027

Master of Science in Artificial Intelligence

University of Jyväskylä · Jyväskylä, Finland

Thesis underway · multi-objective optimization

2021

Bachelor of Science in Computer Science & Engineering

American International University-Bangladesh · Dhaka, Bangladesh

Completed

Toolkit

How I approach the work.

A practical mix of product thinking, systems craft, and curiosity.

SHIP

Production-minded

From zero-downtime data migrations to CI/CD, I care about the details that keep software dependable after launch.

EXPLORE

Applied AI

I build practical RAG and LLM experiences that connect retrieval, infrastructure, and a useful end-user workflow.

CONNECT

Team multiplier

I enjoy turning hard-won engineering lessons into clear standards, reusable components, and better collaboration.

Technical toolkit

The tools behind the work.

A practical stack spanning product interfaces, distributed services, cloud infrastructure, and applied AI.

Languages

TypeScriptJavaScriptPythonGo

Backend / API

Node.jsExpress.jsGraphQLRESTDjango

Frontend

React.jsNext.jsVue.jsTailwind CSS

Data stores

PostgreSQLMongoDBRedisChromaDBDynamoDB

Cloud / DevOps

AWSDockerCI/CDVercelNginx

AI / ML

RAG PipelinesLangChainCrewAITensorFlowLLMs integration

Contact

Have a hard problem?Let's make it useful.

I'm always interested in thoughtful product teams, ambitious engineering challenges, and applied AI that earns its place in the workflow.

salman.prottoy@gmail.com