About
I like solving problems that don't have Stack Overflow answers.
I'm Aamir. I spend most of my time writing Python and wondering why my queries are slow.
I got into engineering because I wanted to understand how things work not just use them. When something breaks, I need to know why it broke, not just how to patch it.
These days I'm focused on two things: building data systems that don't fall over under load, and figuring out how to make AI models actually useful for specific problems.
The path here
Now
Deep in AI infrastructure
Spending most of my time figuring out how to make LLMs useful for real problems. Currently obsessed with quantization techniques and why most fine-tuning tutorials don't work at scale.
2020 – 2023
The data pipeline years
Worked on a vehicle data platform that ingested 200GB daily. Learned the hard way that PostgreSQL isn't the answer to everything. Rebuilt the entire search infrastructure twice.
2017 – 2019
Learning to build
Cut my teeth on Django projects for US clients. Made every mistake in the book N+1 queries, missing indexes, deploying on Fridays. Each failure taught me something.
How I approach problems
I prefer boring technology
SQL, Django, Redis. They're not exciting, but they work at 3 AM when I'm asleep. I only reach for new tools when I've hit a real wall with proven ones.
I need to understand the 'why'
Can't just use an API without understanding what's happening underneath. That's why I'm relearning linear algebra using embeddings without knowing the math felt wrong.
Ship first, optimize later
I've learned that the best way to find problems is to put something in front of real data. Most of my best insights came from production failures, not planning.
Credentials
Experience
7+ Years
Specialization
Backend & AI Systems
Location
United Kingdom
Languages
Python, TypeScript, SQL
Areas of expertise
Current setup
OS
Linux / MacOS
Editor
VS Code with Vim motions
Reading
Designing Data-Intensive Applications