Eman Munir
Chaudhary
AI Engineer & Data Scientist
I build intelligent systems using cutting-edge AI technologies. Specialized in RAG pipelines, machine learning architectures, and enterprise AI solutions. Team Lead at AB Ark, driving innovation in AI engineering.
I'm an AI Engineer and Data Scientist currently serving as Team Lead AI at AB Ark Private Limited. With a passion for building intelligent systems, I specialize in developing sophisticated AI solutions that solve real-world problems.
My expertise spans Retrieval-Augmented Generation (RAG) systems, multi-modal AI, machine learning architectures, and enterprise-scale deployments. I've led the development of production-grade AI systems including Confluence knowledge chatbots, speech analytics pipelines, and emotion analysis systems.
Currently pursuing a Bachelor's degree in Data Science from FCIT, University of the Punjab with relevant coursework in Deep Learning, Machine Learning, MLOps, and Computer Vision. Beyond technical work, I'm passionate about competitive programming and community leadership as President of the Sports Society.
Lahore, Pakistan
Driven to build AI systems that empower people and solve real-world challenges.
Team Lead AI / AI Engineer
AB Ark Private Limited
- •Led development of multi-modal AI systems including RAG pipelines, virtual try-ons, and emotion analysis
- •Built scalable retrieval-augmented generation (RAG) systems integrating FAISS, Chroma, and Pinecone for contextual chat
- •Designed and deployed enterprise AI assistants integrated with Confluence and Microsoft Teams using Google Vertex AI
- •Engineered speech-to-text analytics pipelines using FastAPI, AssemblyAI, and GPT models for call quality and compliance analysis
- •Managed AI team workflows, reviewed code, and guided project deployments to AWS and GCP
Multi-LLM RAG Framework
Engineered a modular Retrieval-Augmented Generation framework supporting Simple, Advanced, and Hybrid retrieval chains with multi-LLM orchestration.
Dynamic LLM selection between OpenAI and local models for optimal accuracy and latency
Vertex AI Confluence RAG for Microsoft Teams
Built a production-grade Confluence Knowledge Base Chatbot using Google Vertex AI embeddings and Gemini models for enterprise-scale contextual QA.
Automated document synchronization via cron-based jobs ensuring real-time retrieval context
South Africa Call-Bot
Developed a FastAPI-based speech intelligence microservice for end-to-end call analytics and QA with speaker diarization and GPT evaluations.
Asynchronous pipelines with scalable ingestion and comprehensive conversation scoring
Therapy Emotion Analysis System
Designed a multi-modal system analyzing emotion and posture in therapy sessions using computer vision and speech intelligence.
Supports both offline and real-time analysis providing clinicians with quantitative behavioral insights
Virtual Try-On System
Built a virtual try-on pipeline using Stable Diffusion and ComfyUI workflows for realistic human-clothing compositing.
Lightweight, production-ready model for scalable fashion content creation and e-commerce demos
INOVO-OCR Pipeline
Developed a multilingual OCR pipeline with custom GPT post-processing and trained YOLOv8 model for receipt detection.
Advanced pre-processing and layout analysis for improved text recognition on complex receipts