Oh.
You're here.
Give me a second.
I was doing something.
|
I'm Ayesha.
I build software, mostly backend stuff, and lately I've been getting curious about how AI actually works.
|
An AI-powered career assistant built around RAG. The idea:
And then:
That's what got me interested in retrieval, embeddings and semantic search. Built with
|
A job discovery backend. Mostly APIs, databases and NLP. The sort of project that starts with:
It wasn't. Built with
Focus REST APIs · NLP · database design |
A Postgraduate Academic Management System backend. Built to manage postgraduate students, supervisors and thesis workflows. Handles Student eligibility · supervisor requests · thesis proposals · progress tracking · defense workflows · role-based access Roles
Built with
|
I'm pretty new to it.
Right now I'm learning the fundamentals and trying to understand what's actually happening rather than immediately reaching for a library.
For example:
model.fit(X, y)I'd like to know what that line is actually doing.
Currently working through:
Linear Regression · Gradient Descent · Classification · Neural Networks
Very much a work in progress.
RAG · LangChain · LangGraph · LLM workflows · Agents
|
REST APIs · Authentication · JWT · RBAC · Business Logic |
||||||||
|
Schema design · Queries · Relationships · Migrations |
Exploring:
|
||||||||
|
|
|
|
REPOSITORIES |
HACKATHON FINALS |
TOP 100 |
|
|
|
Solvio AI Hackathon 2025 |
|
CodeSprint — Inter-IUT Hackathon |
|
SUST CSE Carnival Hackathon |
|
IUT 12th ICT Fest — Agentic AI Hackathon |
Three finalist finishes.
One Top 100 placement.
And apparently a growing collection of certificates.
┌─ CURRENTLY OPEN ───────────────────────────────┐
│ │
│ [01] Machine learning fundamentals │
│ [02] RAG / LLM rabbit holes │
│ [03] One extremely fast paced thriller │
│ │
└────────────────────────────────────────────────┘
One of these is going significantly better than the others.
I will not tell you which one.
Okay.
So either:
A) something here caught your attention.
B) you clicked by accident.
C) you were intrigued by my profile pic.
D) you also have too many tabs open.
Honestly, all four are valid.
If you want to talk about something I'm building, something you're building, or how I ended up reading some extremely niche historical fact at 2 AM—
You may now return to whatever you were supposed to be doing.