Open to opportunities

zaheer abbas

@zaheerabbas1

Message

Senior AI/ML engineer building production LLM, RAG, and multi-agent systems.

Pakistan

Message

What I'm looking for

I want to build production LLM/RAG and multi-agent systems end-to-end—owning deployment, latency, and measurable outcomes—alongside strong MLOps practices (Docker/Kubernetes, monitoring, CI/CD) and fast, iterative engineering.

I’m an accomplished AI/ML Engineer with 3+ years of production experience architecting and deploying scalable, enterprise-grade intelligent systems that generate measurable business outcomes. I specialize in LLM fine-tuning (LoRA/QLoRA/RLHF), Retrieval-Augmented Generation (RAG), multi-agent orchestration, and real-time computer vision—backed by consistent delivery of 89–92% model accuracy and sub-second, production-grade performance.

I build end-to-end solutions across the modern AI stack, from PyTorch and Hugging Face Transformers to vector databases (Pinecone/Chroma/Weaviate/FAISS), MLflow tracking, and Docker/Kubernetes deployment. In recent work, I orchestrated a 40+ agent ecosystem (CrewAI/AutoGen) that reduced manual task-completion time by up to 60%, deployed a YOLOv8 pipeline with sub-100ms inference latency, and built autonomous RAG systems achieving sub-200ms query latency at 99.5% uptime.

Experience

Work history, roles, and key accomplishments

Current

AI-Integrated Mobile Engineer

Current

Freelance Consultant

Feb 2024 - Present (2 years 4 months)

Integrated on-device AI inference into Flutter apps using optimized FastAPI backends, reducing data-synchronization latency by 40% via advanced async state management. Implemented secure Firebase Auth/Firestore and delivered quantized-model NLP chat and computer-vision features for apps serving 500+ active users at 99.9% availability.

Flutter fastAPI Firebase Authentication Firestore TFLite PyTorch Quantization On Device Inference REST APIs Async States

Current

AI System Full-Stack Developer

Current

Freelance Consultant

Sep 2023 - Present (2 years 9 months)

Built scalable REST APIs (Flask/FastAPI) handling 10,000+ daily requests with average response times under 200ms, and implemented Redis caching improving retrieval by 45%. Established CI/CD with GitHub Actions and Docker to cut deployment cycles by 30% while integrating ML/RAG pipelines into user-facing web applications.

fastAPI Flask REST API Development Redis Docker GitHub Actions Model Versioning

Current

Senior AI/ML Engineer

Current

Freelance Consultant

Jan 2023 - Present (3 years 5 months)

Architected and deployed a multi-domain AI platform with LLM-powered agents, RAG, and real-time computer vision, processing 60,000+ records and 5,000+ images monthly at 99.5% uptime. Delivered LoRA/QLoRA + RLHF fine-tuning, agentic orchestration (40+ agents), and a YOLOv8 vehicle detector with 87% mAP and sub-100ms edge inference latency.

LoRA Q LoRA Fine Tuning Faiss AutoGen Vector Search YOLOv8 TensorRT

Education

Degrees, certifications, and relevant coursework

COMSATS University Islamabad

Bachelor of Science, Computer Science

Bachelor of Science in Computer Science at COMSATS University Islamabad, with coursework covering machine learning, deep learning, NLP, and distributed computing. Expected graduation is 2026.