Skip to content
Botmer International®
Botmer talent profile

Hamza Sarfraz

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.

4+ years
dated professional AI experience
Python & FastAPI
application engineering stack
RAG & model training
technical specialization

Islamabad, Pakistan

Meet the engineer

AI engineering with production depth

Where Hamza adds value

  • Led a twelve-person team at Turing across alignment, supervised fine-tuning, RLHF workflows, evaluation frameworks, and training dataset quality controls.
  • Developed Tensor Labs chatbots and question-answering applications with RAG, LangChain, Pinecone, ChromaDB, and pgvector retrieval infrastructure for NLP products.
  • Built Flask and FastAPI backends, ETL pipelines, SQL optimizations, and AWS or GCP model deployments for scalable application integration.
  • Delivered customer segmentation, sales forecasting, OCR receipt processing, and YOLO object detection through independent professional contracts before joining Tensor Labs.
Explore professional experience
Core competencies

AI systems grounded in software engineering

01

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.

  • RAG
  • LangChain
  • Pinecone
  • ChromaDB
  • pgvector
  • NLP
02

Model training and evaluation

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.

  • SFT
  • RLHF
  • PyTorch
  • Hugging Face
  • A/B evaluation
  • Dataset engineering
03

Backend and cloud delivery

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.

  • Python
  • Flask
  • FastAPI
  • SQL
  • AWS
  • GCP
04

Applied machine learning

Delivers segmentation, forecasting, document processing, and image detection applications, combining K-Means, OCR, YOLO, TensorFlow, and PyTorch with technical mentoring and project delivery responsibilities.

  • K-Means
  • OCR
  • YOLO
  • TensorFlow
  • PyTorch
  • Forecasting
Experience

From models to product delivery

Aug 2025 – Feb 2026Turing

AI Lead

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.

  • Developed supervised fine-tuning notebooks and RLHF pipelines.
  • Directed training data synthesis for Linux and browser computer-use trajectories.
  • Built A/B evaluations and adversarial hallucination testing.
  • Established model governance and quality-control workflows.
Dec 2022 – Aug 2025Tensor Labs

Machine Learning Engineer | AI Engineer

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.

  • Built RAG chatbots and question-answering applications.
  • Developed Flask and FastAPI services for web and mobile integration.
  • Deployed models on AWS and GCP.
  • Implemented ETL pipelines and optimized SQL queries.
Dec 2021 – Nov 2022Independent Professional Contracts

Independent Professional

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.

  • Built customer segmentation and sales forecasting reports.
  • Automated receipt scanning and parsing with OCR.
  • Developed YOLO detection and a desktop point-of-sale application.
Selected work

AI systems built for practical use

Highlights from professional projects, product delivery, and specialist work.

Turing

Model alignment and training workflows

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.

  • Python
  • PyTorch
  • Hugging Face
  • RLHF
Tensor Labs

RAG and question-answering applications

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.

  • LangChain
  • FastAPI
  • Pinecone
  • AWS
Independent contracts

Document and predictive automation

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.

  • K-Means
  • OCR
  • YOLO
Education & professional focus

A foundation for practical delivery

Education & qualifications

Bachelor of Computer Science, Artificial Intelligence

Bahria University Islamabad

Jan 2018 – Dec 2022

Professional focusEnd-to-end AI application delivery

Builds retrieval applications from data preparation and embeddings through backend APIs and cloud deployment, drawing on Tensor Labs NLP products and independent automation projects.

Professional focusTraining quality and technical leadership

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.

Talk to Botmer

See if Hamza fits your AI engineering team

Share your requirements, current challenges, and delivery goals. Botmer will review the brief and help arrange the next step.

Request a talent match