Prithiv Sakthi

prithiv sakthi

AI enthusiast with a focus on data-driven solutions, databases and generative models. Proficient in leveraging advanced AI techniques to derive insights, solve complex problems, and drive innovation. Dedicated to continuous learning and pushing the boundaries of AI technology.

Download Resume

My Skills & Work

ABOUT

I'm passionate about exploring the exciting field of Machine Learning, Data-driven solutions and its applications, particularly in the realms of AI, databases, and generative models. As an AI enthusiast, I focus on data-driven solutions and leveraging advanced AI techniques to derive insights, solve complex problems, and drive innovation. I'm dedicated to continuous learning and pushing the boundaries of AI technology. For more visit strangerzone.world

  • portfolio

    DATA BASE

    Meta Database Certification

  • portfolio

    DATA SCIENCE

    Data Science Boot Camp

  • portfolio

    AI / ML

    Machine Learning With Python x R Lang

  • portfolio

    PROMPT ENGINEERING

    AI Bootcamp (2024)

  • portfolio

    SDXL / COMFY UI

    Stable Diffusion x Comfy UI Model

  • portfolio

    DATA ANALYTICS

    Google Data Analytics-coursera

Recent Activities and Work

PROJECTS & SUBMISSION

  • STABLE IMAGINE

    LoRA (Low-Rank Adaptation) models enhance the capabilities of Stable Diffusion by providing specialized styles and characteristics. These models adjust the base model’s weights to generate images in specific styles or themes. In our application, we integrate several LoRA models, each designed to capture different artistic elements. Checkout the space here Stable Imagine Space.

    Read more Working
  • SAVE WEB AS ZIP

    Download a landing page, full website, or any page absolutely for free. Add your site's url to the input box and click «Save» button to get the archive with all files. web2zip allows you to download a page of website and compress all file to a ZIP file.

    Read more Working
  • MASTER GPT

    Master Gpt is a concept model that includes a base chatbot, fast-generating chatbots, Image GPT, SmartSearch, Web2Zip, Media Downloader. These features are designed to improve the performance of GenAi with Mistrals 7B instruct v0.1 and newer v0.3, as well as yt-dlp for media assistance. To generate & use th master gpt, you can visit the spaces on Hugging Face.

    Read more Working
  • IMAGE GEN DEMO

    An AI system that can create realistic images and art from a description in natural language using the RealVisXL model via pretrained DiffusionPipelines. These high-resolution images can be used for productivity and can be generated by visiting the spaces on Hugging Face.

    Read more Working
  • EHRM MODELS

    AI-powered EHRs enable convenient patient records management with dictation, converting physical medical records into digital format, suggesting relevant input data, and more. AI-enabled medical billing. AI algorithms can detect insurance fraud, find errors in submitted claims, or automate the medical coding process. Based on HIMSS Health Foundation. ' Tech 4 Health ' .

    Read more Working
  • INSTRUCT CHATBOT

    Instruct Chatbot is a multi-model chatbot named Dorado, Hercules, and Lepus. It is used with inference clients and has a hierarchy of gradual, fast inference from the model (Mixtral, MistralAI instruct versions 8X7B, 7Bs). To generate, you can visit the spaces on Hugging Face. https://huggingface.co/docs/hub/spaces-config-reference.

    Read more Working
  • IMAGEGPT XL

    Image GPT XL is the GPT model for image inference that analyzes and generates results based on the inferred using the pretrained model Microsoft/Phi-3-vision-128k-instruct AutoModelForCausalLM. To generate image inferences, you can visit the spaces on Hugging Face.

    Read more Working
  • Get Started

    MAGIC ERASER TOOL

    Magic Image Eraser is a powerful tool that allows you to remove unwanted objects or people from an image. It uses advanced image processing techniques to identify and remove the unwanted elements from the image, leaving behind a clean and clear image. Once the unwanted elements are separated, the tool uses image interpolation and other techniques to fill in the gaps left behind, resulting in a clean and clear image.

    Read more Working
  • SIGHT SENSE

    A robust sign language recognition system was developed by harnessing the advanced capabilities of the You Only Look Once (YOLO v5) model. This state-of-the-art approach enabled precise real-time object detection, leveraging both TensorFlow (TF) and PyTorch for model development and training. Additionally, a store management component was seamlessly integrated to expand the project's utility.

    Read more Working

Course Work

STACKS

Tech Stacks🚀: AWS ec2, Vercel, GitHub Pages, Render, MySQL, SQLite, Keras, Matplotlib, Plotly, scikit-learn, TensorFlow, Scipy, PyTorch, NumPy, Pandas, mlflow, Power BI, Canva, Blender, Anaconda, FastAPI, Flask, Python3, R, Gradio, Streamlit, Hugging Face.

  • portfolio

    ChatBot Gradio⚡

    Stranger AI ChatBot

  • portfolio

    Rabbit TTS

    Text-to-Speech

  • portfolio

    HEALTHCARE INFORMATICS

    Information System - MediSearch

  • portfolio

    Medical Health Bot

    Mistral-Maturity-LLM

Work & Read more

MARKDOWN BLOG-CARDS