Retrieval and NLP applications
Builds chatbots and question-answering systems using RAG, LangChain, vector databases, embeddings, data pipelines, and deployed APIs, connecting language models to practical web and mobile applications.
AI Engineer & Technical Lead
AI engineer and technical lead with four years of dated professional experience developing language model applications, retrieval systems, and backend services. He has led alignment and training workflows, built NLP products at Tensor Labs, and delivered independent data projects. His stack combines Python, FastAPI, PyTorch, cloud deployment, and vector retrieval.
Islamabad, Pakistan
Builds chatbots and question-answering systems using RAG, LangChain, vector databases, embeddings, data pipelines, and deployed APIs, connecting language models to practical web and mobile applications.
Develops supervised fine-tuning and RLHF workflows, training datasets, A/B evaluations, adversarial testing, and quality controls using Python, PyTorch, and Hugging Face tooling for language models.
Builds Python services with Flask and FastAPI, develops ETL workflows and SQL retrieval, and deploys models to AWS or GCP environments supporting availability and fault tolerance.
Delivers segmentation, forecasting, document processing, and image detection applications, combining K-Means, OCR, YOLO, TensorFlow, and PyTorch with technical mentoring and project delivery responsibilities.
Led a twelve-person team on language model alignment and training, developing fine-tuning and RLHF workflows, computer-use datasets, evaluation protocols, quality controls, and Python-based data-to-training infrastructure for research delivery.
Built and deployed machine learning and NLP applications, connecting retrieval-based chatbots, vector storage, ETL, SQL optimization, backend APIs, cloud model deployment, workflow automation, and engineering mentorship for client systems.
Delivered independent data and software projects spanning K-Means customer segmentation, sales forecasting, OCR receipt parsing, YOLO detection, desktop point-of-sale software, and game applications across multiple target platforms.
Highlights from professional projects, product delivery, and specialist work.
Led supervised fine-tuning and RLHF pipeline development alongside computer-use training data engineering. Built A/B evaluations and adversarial testing to assess hallucinations, with Python, PyTorch, and Hugging Face workflows supporting dataset quality and training readiness.
Designed NLP applications using retrieval augmentation, LangChain, and vector databases including Pinecone, ChromaDB, and pgvector. Connected ETL and SQL data workflows to Flask or FastAPI backends and deployed machine learning services across AWS and GCP.
Delivered customer segmentation with K-Means, sales forecasting reports, OCR receipt scanning and parsing, and YOLO image detection. These independent projects combined applied machine learning with custom software delivery, including a desktop point-of-sale system for book supply.
Bahria University Islamabad
Jan 2018 – Dec 2022
Builds retrieval applications from data preparation and embeddings through backend APIs and cloud deployment, drawing on Tensor Labs NLP products and independent automation projects.
Leads model training and alignment workflows through dataset engineering, evaluation protocols, quality controls, and mentoring, supported by Python infrastructure and a twelve-person technical leadership appointment.
Share your requirements, current challenges, and delivery goals. Botmer will review the brief and help arrange the next step.