Affan
Mehmood

• Built enterprise-grade AI products using React.js (frontend) and Flask (backend)
• Worked on AANI — Advanced Artificial Negotiation Intelligence platform enabling agents to negotiate using AI-driven insights
• Integrated AARI with Salesforce to execute complex queries and dynamically generate analytical tables and charts
• Participated in system design, feature development, and performance optimisation for production deployments

• Delivered 8+ production projects across frontend, backend, ML pipelines, and cloud deployment
• Built a sales coaching platform using React and FastAPI with real-time and post-meeting analytics on video and audio data
• Designed scalable microservice architecture with FastAPI and Ray to support ML inference workloads
• Built and deployed an AI-powered apparel generation pipeline using Stable Diffusion and DALL·E on GCP with Nginx
• Developed OnlyAI, a unified platform integrating 15+ state-of-the-art NLP and Computer Vision techniques
• Created video forensics toolkit with object tracking and face extraction using YOLOv4, DeepSort, and DSFD
• Designed an ensemble ML model combining clustering, classifiers, and regressors — improving F1 score by 12%
• Fine-tuned Google Pegasus on custom datasets to achieve under 50% plagiarism for academic text generation

• Developed full-stack web applications using React, Node.js, and MySQL while completing undergraduate studies
• Built a Reddit-style social platform with real-time features using WebSockets
• Delivered an automatic transcription MVP using Azure Cognitive Speech Services
• Created a hotel booking startup POC used for investor demonstrations

University of Wollongong, New South Wales

Air University, Islamabad