Ehsanur Rahman Rhythm
Software Engineer & Researcher · Dhaka, Bangladesh
I'm a Software Engineer and AI Researcher with four years of experience across full stack product development, systems integration, and applied machine learning research. I'm currently a Software Engineer at iF Returns, a Spain-based returns management platform, and a Product Engineer at DeliveryHobe, a peer-to-peer logistics platform based in Bangladesh.
Alongside my engineering work, I conduct AI accessibility research at BRAC University, with published work in indoor navigation, obstacle detection, sentiment analysis, and multilingual NLP. I have multiple peer-reviewed conference publications and an MSc thesis on multimodal retrieval-augmented generation for Bangla Question Answering.
I hold an MSc in Software Engineering from the University of Dhaka and a BSc in Computer Science from BRAC University. My industry experience spans serverless AWS infrastructure, GraphQL APIs, NestJS and Next.js applications, and e-commerce integrations across Salesforce Commerce Cloud and Shopify.
Currently
- Working on
- Customer returns dashboard at iF Returns and P2P logistics platform features at DeliveryHobe
- Location
- Dhaka, Bangladesh (UTC+6)
Experience
NSBH is a B2B trading credit platform that pays suppliers upfront and lets buyers purchase now and pay later through flexible, automated terms.
Reverse logistics platform improving the customer returns experience for e-commerce businesses across Europe.
Peer-to-peer logistics platform connecting senders with delivery partners across Bangladesh.
AI accessibility research group — enhancing university life for blind students.
Open-source civic tech platform enabling community-driven action for citizens in Bangladesh.
Early-stage startup building local SaaS products and client web solutions.
Research
Deep hierarchical networks for sentiment analysis of restaurant reviews from food apps
Md Humaion Kabir Mehedi, Fahmid Al Farid, Ehsanur Rahman Rhythm, Farhin Rahman, Khan Md Hasib, Jia Uddin, Sarina Mansor
Scientific Reports · Nature · 2025
DOI: 10.1038/s41598-025-23856-5Multimodal Retrieval-Augmented Generation for Bangla Question Answering
Ehsanur Rahman Rhythm
IIT, DU · IIT, DU · 2025
Automatic Subtitle Generation for Bengali Multimedia Using Deep Learning
Ehsanur Rahman Rhythm, Shafakat Soroar Arnob, Rajvir Ahmed Shuvo, Sifat E Jahan, Annajiat Alim Rasel
BRACU · BRACU CSE · 2023
http://hdl.handle.net/10361/23554Education
Master of Science in Software Engineering
Grade: 3.44
Thesis: Multimodal Retrieval-Augmented Generation for Bangla Question Answering
Bachelor of Science in Computer Science
Grade: 3.70
Thesis: Automatic Subtitle Generation for Bengali Multimedia Using Deep Learning
Awarded Dean’s List honours 4 times for maintaining top-tier academic performance. Named to the Vice Chancellor’s List, recognizing excellence in Spring 2021.
Beyond the code
I'm a big Formula 1 fan — the strategy, the engineering margins, and the sheer pace of it all. I try not to miss a race weekend. When I'm not watching F1, I play Rocket League competitively and occasionally pick up PUBG PC.
I'm married, a proud cat owner, and actively working on keeping screen time balanced with real life. I type at 142 WPM (100% accuracy), which probably says something about me.
Connect
The best way to reach me is via email. I'm also on most of the usual platforms.