Peshawar, Khyber Pakhtunkhwa, Pakistan
🚀 Senior AI Engineer | Python, NLP, FastAPI, LangChain, RAG, LLMs | 5+ Years Creating GenAI Applications I’m a results-driven AI Engineer with 5+ years of experience designing, building, and scaling AI-powered applications. My expertise lies in Natural Language Processing (NLP), Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and backend engineering, where I bridge cutting-edge AI with production-ready systems. At TalentPop, I’ve led the end-to-end development of advanced conversational AI platforms, semantic search engines, and RAG-based applications using LangChain, OpenAI APIs, Hugging Face embeddings, and Qdrant vector databases. From property recommendation systems to customer service analysis platforms, I’ve consistently delivered real-time, scalable, and high-impact solutions that improve decision-making and user experience. Previously at Afiniti, ISSM.ai, and NCAI-AIH, I built expertise in chatbot frameworks (Rasa, Nest.js), multimodal AI, enterprise APIs, and data pipelines. With hands-on experience in Docker, AWS, PostgreSQL, and modern Python frameworks (FastAPI, Flask, Sanic), I combine deep AI knowledge with robust backend engineering skills. 💡 Key Highlights: Built RAG pipelines integrating PDFs, URLs, and vector stores for accurate Q&A. Designed scalable AI APIs with FastAPI and SQLAlchemy powering real-time analytics. Developed ML/NLP models for sentiment analysis, grammar correction, and classification. Created asynchronous PDF/image extraction pipelines with MongoDB storage. Delivered enterprise chatbots and AI workflows using LLMs, embeddings, and fine-tuning. 🔑 Core Skills: Python • FastAPI • NLP • LLMs • RAG • LangChain • Hugging Face • Qdrant • PostgreSQL • Docker • AWS • CI/CD • Semantic Search • GenAI Applications I’m passionate about pushing the boundaries of AI, from research to deployment, helping companies build smarter, faster, and more human-like AI systems. Always open to opportunities where AI innovation meets real-world impact.
At TalentPop, I’ve spearheaded the design and development of AI-powered platforms that combine NLP, LLMs, LangChain, and vector databases to solve real-world business problems. Property Recommendation System – Built a conversational AI solution using OpenAI, LangChain, and Qdrant that delivers personalized property recommendations with dynamic filtering (location, price, rooms, etc.), providing users with fast, intuitive, and scalable search experiences. Customer Service Conversation Analysis – Developed a FastAPI-based analytics platform leveraging OpenAI APIs and PostgreSQL to grade and evaluate customer interactions. Delivered actionable insights on trends and conversation quality, boosting client engagement and operational decision-making. Web Applications with NLP & ML – Created sentiment analysis and grammar correction engines using NLTK, spaCy, HappyTransformer, integrated with PostgreSQL and FastAPI to provide real-time feedback and improve user engagement. RAG Applications – Built robust question-answering systems using LangChain pipelines, Qdrant, and OpenAI models. Integrated PDF/URL processing, similarity search, and retrieval chains for highly accurate responses. PDF & Image Processing – Designed an asynchronous pipeline to extract and filter images/text from PDFs, optimized storage via MongoDB, and implemented bulk insertion for high-performance data handling. Impact: Delivered scalable, high-performance AI systems that streamlined decision-making, enhanced customer experience, and improved adoption of business-critical platforms.
● Reviewed project specifications and designed technology solutions that met or exceeded performance expectations. ● Worked with software development and testing team members to design and develop robust solutions to meet client requirements for functionality, scalability, and performance. ● Coordinated with other engineers to evaluate and improve software and hardware interfaces. ● Represented software applications engineering team during large and complex development projects. ● Offered experience with Typescript, RASA and Python. ● Developed robust, scalable, modular and API-centric infrastructures. ● Worked on a wrapper around the Conversational AI platform to build a scalable product involving multiple services including, docker service, connector service (to communicate with different services). ● Built a chat bot backend in Nest.js that lets users build chat bots using flowcharts. ● Developed a Docker service that instantiates chat bot containers. ● Utilized a modular approach with micro services architecture to minimize downtime and make maintenance simple.
Reviewed project specifications and designed technology solutions that met or exceeded performance expectations. Worked with software development and testing team members to design and develop robust solutions to meet client requirements for functionality, scalability, and performance. Coordinated with other engineers to evaluate and improve software and hardware interfaces. Represented software applications engineering team during large and complex development projects. Offered experience with Typescript, RASA and Python. Developed robust, scalable, modular and API-centric infrastructures. Worked on a wrapper around the Conversational AI platform to build a scalable product involving multiple services including, docker service, connector service (to communicate with different services). Built a chat bot backend in Nest.js that lets users build chat bots using flowcharts. Developed a Docker service that instantiates chat bot containers. Utilized a modular approach with micro services architecture to minimize downtime and make maintenance simple. Skilled in Large Language Model (LLM) development, including fine-tuning, tokenization, and alignment with human values. Proficient in LLM architecture and sequence prediction, with expertise in integration tools like ChatGPT and Chainlit. Experienced in developing document question answering systems using techniques such as chunking and cosine similarity searches, enabling efficient data retrieval from vector databases and embedding. Specialized in advanced topics like web browsing agents and chatbot UI integration, with a strong foundation in LLM-based technologies for chatbot interfaces, document conversations, and web research.
** Khushhaali Microfinance Bank Ltd Chatbot: Khushhaali is amongst the best microfinance banks. * Responsibilities: - Building middleware for chatbot and API integrations. - API integrations with social media channels i.e. Facebook, WhatsApp - Using state-of-the-art machine learning frameworks such as RASA & Dialogflow to train chatbots. - Using Natural language processing for cleansing the dataset. ** HabibMetro Bank Chatbot: A Pakistani subsidiary of Swiss bank Habib Bank AG Zurich. * Responsibilities: - Building middleware for chatbot and API integrations, in Flask, Sanic - API integrations with social media channels i.e. Facebook, WhatsApp - Building an Automated Roshan Digital Account (RDA) Chatbot, for registration of users using multiple entities and slots and forms, in English and Roman Urdu as well. - Build a comprehensive FAQs chatbot of RDA. ** Allied Bank Chatbot: The fifth largest commercial bank in Pakistan * Responsibilities: - Building an Automated Roshan Digital Account (RDA) Chatbot, for FAQs - Build a comprehensive FAQs chatbot of Allied Bank users specifically. Other Responsibilities: ● Designed and developed Chatbots for various channels (web, Facebook Messenger, SMS, etc.), and worked in CI/CD, Git, unit-testing and source code management, all while gaining knowledge of Natural Language Processing, Machine Learning algorithms, models, and principles. ● Worked on deployment for the projects on AWS services i.e. EC2, S3, IAM, and Amazon Sagemaker. ● Created APIs using Python (Sanic framework) for the hardware to send data and communicate with the front-end. ● Gained deep knowledge of most commonly used cloud services such as AWS, Azure, Google Cloud, etc. ● Core Technologies worked on included Python, NLP, NLU, Speech Recognition and frameworks like Django, Sanic, Flask, etc.
• Researched on Artificial Intelligence based Multi-modal Medical Image Analysis, which involves the Brain MRI images for Tumor Detection. • I have performed Sentiment Analysis through Deep Learning using IMDB dataset. • Another research which I was working on is LSTM-Based ECG classification. • Machine Learning with R: Perforned Supervised vs Unsupervised Learning, looked into how Statistical Modeling relates to Machine Learning, and do a comparison of each. • Packet Inspection Using Deep Learning: Project on Packet Inspection on Streaming Data to capture the packets over the network using Wireshark and train our model to differentiate between video packets and others.