I am a Computer Science and Engineering student at Shahjalal University of Science and Technology (SUST) and a Software Engineer Intern at Recurse.AI. I specialize in building AI-powered web applications and solving complex algorithmic problems.
- π Education: Final year BSc in CSE at SUST (CGPA: 3.77/4.00).
- π» Current Role: Software Engineer Intern at Recurse.AI.
- π Competitive Programming: 1000+ problems solved. Codeforces Max: 1510.
- π§ Interests: AI/ML, Full-Stack Development, and Speech-to-Speech frameworks.
| Category | Skills |
|---|---|
| Languages | C++, C, Python, Java, JavaScript, SQL |
| Frontend | React.js, Next.js, HTML5, CSS3 |
| Backend | Node.js, Express.js, Spring Boot, Java Servlet |
| Mobile | React Native, Expo |
| Databases | PostgreSQL, MongoDB, MySQL |
| Tools/Cloud | Docker, Git/GitHub, Postman, Prisma ORM, Pinecone |
- Rank 14th β ICPC Dhaka Regional '25
- Rank 20th β ICPC Preliminary '25
- Rank 18th β MU IUPC '25
- Rank 11th β IIUC IUPC '23
- Rank 6th β LU Sylhet Divisional '23
Thryve | Lead Developer
AI-Powered YouTube Content Creation Platform
- Built an automated system for video ideation, voiceovers, thumbnails, and reels.
- Stack: Next.js 15, PostgreSQL, Prisma, Pinecone, OpenAI, SmythOS AI Agents.
TikaSheba | Backend & Mobile Developer
National Vaccination Workflow Automation
- Digitized vaccination workflows with real-time tracking and ML-based forecasting using Meta Prophet.
- Stack: Node.js, Express, MongoDB, Gemini API.
Coder's Platform | Lead Developer
AI-Enhanced Learning Platform
- A collaborative coding environment with personalized AI tutoring.
- Stack: Spring Boot, React, MongoDB.
An End-to-End Speech-to-Speech Framework for Expressive Bengali Accent Conversion using a Novel Parallel Corpus
Jun 2025 β Present
- Developed an S2S framework converting regional Bengali accents to Standard Bengali while preserving speaker identity.
- Created the Bengali Parallel Accent Corpus (BPAC), a novel dataset using a fixed-text protocol for ground-truth pairs.
- Implemented a generative pipeline using Flow Matching Transformers, achieving superior naturalness over Soft-Unit Adaptation.
Investigating the Impact of Fast-Paced Digital Content on Attention and Cognitive Load: A Machine Learning Approach
Mar 2025 β Present
- Applying machine learning techniques to analyze digital behavior patterns and cognitive impacts of social media.
- Investigating behavioral impulsivity and focus using data-driven modeling.
- π₯ Champion - NEUB Hackathon 2025
- π₯ Champion - LUCC Divisional Hackathon 2024
- π₯ 1st Runner-Up - HackTheAI Hackathon 2025
- π₯ 1st Runner-Up - Technocrats V.2 Techathon 2024
- π₯ 2nd Runner-Up - KUET AI & API Hackathon 2025
- π Finalist - Code Samurai 2024 & Therap JavaFest 2023
I'm always open to collaborating on innovative AI or Web projects. Feel free to reach out!
- Location: Sylhet, Bangladesh π§π©
- LinkedIn: unayes-ahmed-khan
- Personal Site: unayes.netlify.app