Open to opportunities

Ahsan Alif

@ahsanalif

Message

Full-stack engineer delivering fast, correct web systems and applied ML privacy research from day one.

Bangladesh

Message

What I'm looking for

I’m looking for a team where I can ship production features quickly, obsess over real performance/correctness, and contribute to privacy-aware ML—while working with strong engineering practices like testing and CI/CD.

I build production-grade full-stack systems with a strong bias for performance and correctness. I’ve deployed three production systems, including a Redis rate-limiter rewrite using an atomic Lua script that fixed a race condition and cut p99 latency 5×, plus a test-isolation bug catch before it could reach production.

On the frontend and backend, I create real-time, user-focused experiences (e.g., Monaco-based collaborative editing with Yjs CRDTs) and design secure data flows with role-based access and database-level protections like PostgreSQL Row-Level Security. I also bring applied ML depth—PyTorch models for multi-task segmentation/classification and federated learning research with differential privacy—so I can contribute beyond web systems while staying grounded in measurable outcomes.

Experience

Work history, roles, and key accomplishments

Federated Learning Research

BRAC University

Jan 2025 - Jan 2026 (1 year)

Conducted final-year research on a federated learning system for remote patient monitoring across 10 hospital nodes without raw patient data leaving devices. Applied differential privacy (DP-SGD) and identified the privacy-utility operating point (epsilon=1.0, delta=1e-5).

Federated Learning Anomaly Detection PyTorch HIPAA GDPR 10 Node Hospital Setup

Co-Founder, Amgo Mango

Amgo Mango

Jan 2023 - Jan 2025 (2 years)

Co-founded a D2C mango e-commerce startup while studying full-time. Scaled operations to 1,000+ customers in Bangladesh with end-to-end ownership of sourcing, logistics, and operations.

Start Up Operations Sourcing Logistics Customer Growth

Brain Tumor Segmentation

Kaggle

Implemented a multi-task U-Net model for BRISC 2025 MRI data to perform both pixel-level segmentation and 3-class tumor classification. Achieved 0.85 Dice and 0.77 IoU for segmentation and 98% classification accuracy.

PyTorch U Net Attention U Net OpenCV scikit learn Dice+BCE Loss AMP Cosine Learning Rate Scheduling Dice IOU

AI Interview Prep Platform

LevelUp.io

Built a full-stack AI interview prep platform with real-time interview rooms, REST API, and role-based access control. Rewrote a Redis rate limiter using an atomic Lua script, reducing p99 latency 5× and improving throughput 2.5× at 50 concurrent connections.

React 19 Express PostgreSQL Redis Socket.IO JWT RBAC Rate Limits Lua Scripting CRDTs Monaco Editor REST APIs

Solvio AI Hackathon

Solvio AI

Advanced to Phase 2 of the Solvio AI Hackathon, selected as top 393 of 3,259 teams.

Robotics Club Junior Exec

BRAC University

Served as Junior Executive for the BRAC University Robotics Club and planned and coordinated technical events across departments.

Event Planning technical coordination

Student Notes Marketplace

NoteBazar

Developed a student notes marketplace with wallet payments, PDF watermarking, and an admin dashboard. Implemented data protection with PostgreSQL row-level security (33 RLS policies, 7 server-side functions) and wrote 112 automated tests.

React 19 Supabase PostgreSQL Vercel PDF Watermarking Testing

Education

Degrees, certifications, and relevant coursework

BRAC University

Bachelor of Science, Computer Science

2022 - 2026

Activities and societies: Relevant coursework: Data Structures & Algorithms, Object-Oriented Programming, Operating Systems, Database Management Systems, Computer Networks, Machine Learning.

Earned a BSc in Computer Science at BRAC University (Feb 2022–Jan 2026), studying topics including Data Structures & Algorithms, OOP, Operating Systems, DBMS, Computer Networks, and Machine Learning.