Beyond the resume, my story.
Where I come from, how I work, and what I'm chasing next.
Hi, I'm Shahriar Khan.
I recently completed my B.Sc. in Computer Science & Engineering at East West University in Dhaka. My focus is production AI, the kind that survives messy real-world data, deploys to edge devices, and answers to actual users. I work across the stack because the boundary between "model" and "product" is fuzzier than most courses suggest.
My Story
My path into AI began the way a lot of practical engineers' paths begin: I wanted to know how the things on my screen actually worked. That curiosity turned into long nights breaking neural networks, then patching them, then breaking them again on slightly different data, until I started understanding the difference between a model that demos well and a model that ships.
At East West University I built systems I'm genuinely proud of: a real-time traffic surveillance platform built in collaboration with the Dhaka Metropolitan Police, an ML-powered dashboard for the national parliamentary election, an explainable-AI tool for image classification, and an AI learning platform with real users. None of these were assignments. They were attempts to learn what real systems look like.
Today I'm focused on the parts hiring managers rarely see: robust data pipelines, reproducible training, edge optimization, and clean APIs. The end result is software that works on day one and still works long after launch.
What I Care About
Continuous Learning
The field moves fast, so I commit to learning something new every week, whether a paper, a framework, or a better engineering practice.
Impact Over Hype
I'd rather solve a real problem with a small model than chase a benchmark. Value first, buzzwords later.
Open Collaboration
The best ideas come from diverse perspectives. I believe in open-source contribution, mentorship, and lifting others up.
Quality & Craftsmanship
Code is read more than written. I take pride in clean, tested, well-documented work, even when no one is grading it.
How I Work
Approach
- Deep work. Long uninterrupted blocks for hard problems.
- Research first. Understand the problem before touching the keyboard.
- Iterate fast. Ship a thin slice, get feedback, sharpen.
- Document. Future me always thanks present me.
Best environment
- Quiet mornings. Most of my best code happens before noon.
- Lo-fi or instrumental music. Background that disappears.
- Whiteboards & pair sessions. Architecture is a conversation.
- Honest review. Critique sharpens work faster than praise ever does.
Beyond Code
Photography
Former Vice President of the Government Science College Photography Club. Storytelling through frames.
Competitive Programming
Member of the EWU Programming Club. ICPC-style contests and weekly problem solving.
Design Tinkering
I designed every iteration of this portfolio myself, and I'm never quite finished.
What I'm Looking For
I'm seeking roles or research collaborations where I can work alongside thoughtful teams on hard ML problems. I thrive in environments that reward learning, experimentation, and shipping. If you're building something at the intersection of research and production, let's talk.
“The best way to predict the future is to build it.”
Alan Kay