Open to opportunities

Khizar Younas

@khizaryounas

Message

I'm building FastAPI and LLM-powered backend services for practical AI products.

Pakistan

Message

What I'm looking for

I'm looking to keep building practical backend and applied AI systems, especially FastAPI services, LLM integrations, and production-focused products where I can deepen my engineering experience.

I'm building FastAPI services at Fly Rank that integrate Groq LLM APIs, including a provider-agnostic summarization endpoint that allows backends to be swapped without changing call sites.

I've implemented database-backed response caching, immediate-retry logic, and per-request token cost logging to reduce redundant API calls, improve repeat-query response times, and monitor LLM spend. Earlier at DecodeLabs AI, I worked in Python on API integrations and reusable backend service components.

I built Nova, an end-to-end AI customer chatbot for Pak Sewing Machines using FastAPI, Groq Llama 3.3 70B, and a custom HTML/JavaScript frontend. I also built the company's React product website with a browsable catalog and responsive interface.

For my final year project, I'm designing AI-IDS, an Android mobile-privacy intrusion detection system using Kotlin, FastAPI, MongoDB, Redis, NetworkX, and PyTorch Geometric to detect anomalous app-permission behavior through graph neural networks.

Experience

Work history, roles, and key accomplishments

Current

Backend AI Engineering Intern

Current

Fly Rank

Jan 2026 - Present (7 months)

Built FastAPI services integrating LLM APIs (Groq), including a summarization endpoint with provider-agnostic architecture. Implemented database-backed response caching, retry logic, and per-request cost logging to optimize performance and control API spend.

fastAPI GROQ LLM APIs MongoDB Redis Python Caching Retry Logic