Open to opportunities

Yousaf Malik

@yousafmalik

Message

Senior Python developer and team lead building scalable GPU-backed backend and AI video inference systems.

Pakistan

Message

What I'm looking for

I want to lead teams building production distributed backend + GPU AI systems—LLM/VLM serving and real-time video inference—where I can own architecture, performance, and quality (reviews/process) while shipping FastAPI services that scale.

I’m a Senior Python Developer and Team Lead specializing in architecting distributed backend systems, multi-tenant SaaS platforms, and GPU-accelerated AI/computer vision pipelines. I lead a 10-person engineering team across backend, AI infrastructure, and frontend codebases, and I’ve helped scale real-time video inference to 160+ concurrent camera streams per GPU node.

I build production-grade services end-to-end—shipping self-hosted LLM/VLM serving infrastructure (vLLM, TensorRT, DeepStream) and FastAPI microservices that handle 10K+ requests per hour. Most recently, I architected and built a GPU-serving microservice from scratch, led migration of the GPU inference pipeline to NVIDIA DeepStream (with a latency-critical C++ extension), and enforced quality through CODEOWNERS-gated reviews and mandatory merge policies.

Experience

Work history, roles, and key accomplishments

Current

Team Lead

Current

Swatitech

Mar 2026 - Present (4 months)

Led a 10-person engineering team across backend, GPU/AI pipeline, edge/central infrastructure, and two frontend codebases. Architected and built a GPU-serving vLLM microservice with a JWT-authenticated FastAPI gateway and led migration to NVIDIA DeepStream 9.0 using a C++ nanobind extension for latency-critical stages.

Team Leadership VLLM fastAPI JWT Nvidia DeepStream C++Code Owners Code Review Pull Requests

Senior Python Developer

Swatitech

Jun 2025 - Mar 2026 (9 months)

Architected a multi-tenant AI video surveillance platform as a distributed system for scaling to 160+ concurrent RTSP camera streams per GPU node. Built a DeepStream inference cascade orchestrator, an async FastAPI backend with Redis Streams consumers, and a multiplexed WebSocket live-feed service integrating RTSP/WebRTC/HLS streaming.

Python fastAPI Distributed Systems Multi Tenant Services RTSP REDIS Streams Faiss WebSockets JSON WebRTC HLS SQLAlchemy JWT RBAC OTP Password Reset Cooldown and Severity Scoring Docker Docker Compose CI CD

Python Developer

AI4LYF Research Institute

Sep 2023 - May 2025 (1 year 8 months)

Designed and implemented Python microservices using FastAPI and async MySQL (SQLAlchemy ORM), handling 10K+ requests per hour. Migrated monolithic systems to microservices using Event Sourcing and AWS-based infrastructure, and integrated Google/Apple in-app purchase payments and Hugging Face ML models into production.

Software Engineer Intern

Arbisoft

Jul 2022 - Sep 2022 (2 months)

Spearheaded development of FeatureDog, an internal feature-request platform using a Django backend and ReactJS frontend with MVP (Model-View-Presenter) architecture. Presented the product to stakeholders at the close of the internship.

Django Model View Presenter Internal Tools Development