Muhammad is available for hire
Hire MuhammadMuhammad Usman Ali
Verified Expert in Engineering
Machine Learning Developer
Lahore, Punjab, Pakistan
Toptal member since May 1, 2026
Bio
Muhammad is a senior AI professional with 8+ years of experience. He designs and deploys scalable solutions across machine learning, deep learning, computer vision, and generative AI. He's specialized in building intelligent systems leveraging large language models (LLMs), retrieval-augmented generation (RAG), agentic AI, and conversational AI (chatbots) to solve complex real-world problems.
Portfolio
Digifloat
Python, WhatsApp, WhatsApp API, Conversational AI, Agentic AI, Webhooks...
Cliquify
Python, RAG Systems, FastAPI, LangChain, PostgreSQL, Vector Databases, Django...
Dough.zone
Python, Computer Vision, Deep Learning, PyTorch, TensorFlow, OpenCV, PostgreSQL...
Experience
- Python - 8 years
- FastAPI - 6 years
- Machine Learning - 5 years
- Deep Learning - 5 years
- Computer Vision - 4 years
- RAG Systems - 4 years
- Generative Artificial Intelligence (GenAI) - 3 years
- AI Chatbots - 2 years
Preferred Environment
Python, Computer Vision, Open-source LLMs, RAG Systems, AI Chatbots, Deep Learning, Machine Learning, Azure Machine Learning
The most amazing...
...projects I've worked involved building AI, LLM, ML, and computer vision solutions.
Work Experience
Senior AI Consultant
2025 - PRESENT
Digifloat
- Led the main WhatsApp bot integration project, designing and implementing an end-to-end conversational AI solution with agent decision logic, memory handling, and fallback flows for production.
- Integrated the WhatsApp Meta Cloud API, including webhook configuration, message routing, session management, and policy-compliant handling of platform rate limits and messaging rules.
- Worked directly with stakeholders and internal teams to define requirements, produce BRD and technical specifications, and support delivery from solution design through deployment and post-launch monitoring.
Technologies: Python, WhatsApp, WhatsApp API, Conversational AI, Agentic AI, Webhooks, LangChain, Large Language Models (LLMs), Generative Artificial Intelligence (GenAI), MLflow, Artificial Intelligence (AI), Machine Learning Operations (MLOps), Retrieval-augmented Generation (RAG), RAG Pipelines
Senior AI Contractor
2023 - 2025
Cliquify
- Engineered a dynamic RAG pipeline that combined vector retrieval with real-time tool use, enabling agents to autonomously query, synthesize, and surface domain-specific recruitment knowledge.
- Designed an agent memory and state management layer using PostgreSQL-backed vector stores, allowing persistent multi-turn context and more adaptive conversational responses.
- Architected and deployed a scalable REST API back end using FastAPI and LangChain/LlamaIndex, enabling production-grade orchestration of tools, knowledge retrieval, and multi-step reasoning workflows.
Technologies: Python, RAG Systems, FastAPI, LangChain, PostgreSQL, Vector Databases, Django, Large Language Models (LLMs), Generative Artificial Intelligence (GenAI), MLflow, Artificial Intelligence (AI), Machine Learning Operations (MLOps), Retrieval-augmented Generation (RAG), RAG Pipelines
Computer Vision Engineer
2021 - 2023
Dough.zone
- Implemented deep learning-based detection and segmentation algorithms on road imagery, enabling automated analysis of surface conditions and improving report generation workflows.
- Applied LLMs to summarize feasibility reports and combined them with statistical road-feature analysis to produce more comprehensive and decision-ready final assessments.
- Led ML developers across the full product lifecycle, from data collection and model training to cloud deployment through REST APIs for production use.
Technologies: Python, Computer Vision, Deep Learning, PyTorch, TensorFlow, OpenCV, PostgreSQL, Artificial Intelligence (AI), Retrieval-augmented Generation (RAG), RAG Pipelines, Neo4j, GraphRAG
Machine Learning Engineer
2018 - 2021
Aimbot studio
- Built a real-time sports video analytics platform using deep learning and computer vision, integrating object detection, multi-object tracking, and classification models to deliver accurate, frame-level insights from live video feeds.
- Applied and fine-tuned machine learning and deep learning models for classification, detection, segmentation, and custom ranking across medical, sports, audio, and geophysical datasets.
- Evaluated and customized state-of-the-art approaches, then deployed production-ready AI solutions across multiple domains, improving reliability and enabling real-world business adoption.
Technologies: Python, Computer Vision, Machine Learning, Deep Learning, PostgreSQL, Django, Artificial Intelligence (AI)
Experience
Large-scale Retrieval Augmented Generation (RAG) System
http://cliquify.meI developed an advanced a retrieval augmented generation (RAG) system, leveraging OpenAI and open-source large language models (LLMs). This innovative system is designed to assist recruiters from leading brands by providing real-time, data-driven answers to their queries.
The RAG system integrates seamlessly with existing data sources, enhancing decision-making and efficiency in the recruitment process. This project showcases cutting-edge AI technology applied to real-world challenges in the recruitment industry.
The RAG system integrates seamlessly with existing data sources, enhancing decision-making and efficiency in the recruitment process. This project showcases cutting-edge AI technology applied to real-world challenges in the recruitment industry.
Automated Crack Detection and Report Generation
The PAVE project is a groundbreaking initiative designed to transform highway maintenance in the United States. This innovative venture focuses on identifying cracks and other forms of distress on highways using advanced road imagery captured through highly sensitive cameras. By automating detection, PAVE aims to optimize the extensive resources annually allocated to road surveys and damage assessments, enhancing efficiency and reducing costs associated with manual inspections.
AI-powered WhatsApp Facility Management Assistant
https://al-ghurair.com/en/al-ghurair-property-managementDeveloped and deployed a production-grade WhatsApp assistant for a facility management business, enabling customers to book, reschedule, cancel, and track service requests directly through WhatsApp.
The solution uses Python, FastAPI, LLM-based agents, Redis session management, and back-end API integrations to guide users through service categories, property selection, issue details, scheduling, and booking confirmation. It also supports structured workflows, validation, error handling, and real-time communication with enterprise facility management systems.
The microservices were containerized with Docker and deployed on Azure Container Apps, with secure environment configuration, logging, scalability, and production monitoring. The assistant reduced manual customer-service effort and provided customers with a faster and more convenient way to manage facility maintenance requests.
The solution uses Python, FastAPI, LLM-based agents, Redis session management, and back-end API integrations to guide users through service categories, property selection, issue details, scheduling, and booking confirmation. It also supports structured workflows, validation, error handling, and real-time communication with enterprise facility management systems.
The microservices were containerized with Docker and deployed on Azure Container Apps, with secure environment configuration, logging, scalability, and production monitoring. The assistant reduced manual customer-service effort and provided customers with a faster and more convenient way to manage facility maintenance requests.
Education
2016 - 2019
Master's Degree in Computer Science
Information Technology University of the Punjab - Lahore, Punjab, Pakistan
2012 - 2016
Bachelor's Degree in Information Technology
Punjab University College of Information Technology (PUCIT) - Lahore, Punjab, Pakistan
Skills
Libraries/APIs
REST APIs, PyTorch, Keras, TensorFlow, OpenCV, WhatsApp API
Tools
Azure Machine Learning, GraphRAG
Languages
Python
Frameworks
Django, LangGraph
Storage
PostgreSQL, Neo4j, Databases
Other
RAG Systems, Generative Artificial Intelligence (GenAI), LangChain, MLflow, Artificial Intelligence (AI), Machine Learning Operations (MLOps), Retrieval-augmented Generation (RAG), RAG Pipelines, Large Language Models (LLMs), Computer Vision, Open-source LLMs, AI Chatbots, Deep Learning, Machine Learning, Data Scientist, Agentic AI, ReAct Agents, FastAPI, Vector Data, WhatsApp, Conversational AI, Webhooks, Vector Databases
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring