hi, i'm Safzan

about me

i'm an ai and machine learning enthusiast with a degree in computer science and engineering (CSE) with a specialization in artificial intelligence and machine learning (AI/ML) and hands-on experience in developing innovative solutions. my passion lies in building intelligent systems that can transform industries and improve user experiences. with a diverse skill set in various programming languages and frameworks, i can quickly adapt to new technologies and deliver high-quality, impactful results.

i have successfully contributed to projects involving generative ai and retrieval-augmented generation (RAG) for large language models (LLMs), systems. my work has focused on developing and fine-tuning ai models to enhance performance and accuracy, particularly in real-world applications. i'm dedicated to continuously learning and growing as a developer, staying current with the latest advancements in ai and related technologies.

in my free time, i enjoy gaming, working on my art, and exploring new ways to use ai to allow for easier self-expression through art.

i'm very much excited to further my expertise in generative ai, machine learning, and other modern technologies, driving innovation and making meaningful contributions to the field.

my links

projects

final year mini project

license plate detection and recognition using YOLOv8 and easyOCR

this project is a part of our final year mini project for computer science engineering. the goal of this project is to design and develop a system that can detect and recognize license plates in videos using yolov8 and easyocr.

  • license plate detection: using YOLOv8, a real-time object detection system, to detect license plates in videos.
  • license plate recognition: using easyOCR, an optical character recognition system, to recognize the text in the detected license plates.

final year major project

offline, open-source rag

retrieve, augment, and generate with open-source large language models

this project provides a self-contained, offline platform for retrieval augmented generation (rag) using open-source large language models (llms). with this project, you can ingesting files and leverage the power of llms without relying on third-party services or exposing sensitive data.

  • offline embeddings & llms support: no need for openai or other external services. our platform uses open-source llms, ensuring your data stays within your network.
  • support for multiple sources: ingest files from various sources, including:
    • local files: upload and process files from your local machine.
    • github repos: ingest files from github repositories.
    • websites: retrieve and process web pages.
  • streaming responses:enjoy real-time responses from the model, enabling a more interactive and conversational experience.
  • conversational memory: the model maintains context and understanding throughout conversations, allowing for more coherent and relevant responses.

benefits:

  • maintain data confidentiality: by keeping data within your network, you minimize the risk of data breaches and ensure compliance with data protection regulations.
  • cost-effective: no subscription fees or usage charges. run the model on your own hardware or cloud infrastructure.
  • flexibility: integrate with your existing infrastructure and workflows.

image comparator

a web-based tool for image manipulation tasks

image comparator is a versatile web-based tool for various image manipulation tasks. it provides a user-friendly interface for comparing, compressing, resizing, cropping, and flipping images.

key features:

  • image comparison: compare multiple images side by side and download the combined result.
  • image compression: compress both JPG and PNG images to reduce file size.
  • image resizing: resize images by specifying width, height, or scale percentage.
  • image cropping: crop images with various aspect ratio options.
  • image flipping: flip images horizontally or vertically.
try it live!

try my projects live!

full chat webui

experience the power of llms with a full chat ui web demo. engage in natural conversations while keeping your data safe and secure.

  • responsive design: enjoy a seamless experience across desktop pc, laptop, and mobile devices.
  • progressive web app (pwa) for mobile: enjoy a native app-like experience on your mobile device.
  • full markdown and LaTeX support: elevate your llm experience with comprehensive markdown and LaTeX capabilities, with full support for programming languages like c and python.
  • multi-models conversations: effortlessly engage with various models simultaneously, harnessing their unique strengths for optimal responses. enhance your experience by leveraging a diverse set of models in parallel, combine their responses to get something refined.
try it live!

blog