光

About

ʕ•ᴥ•ʔ

Hi I’m Manodeep - An Aspiring ML Engineer

Welcome to my first blog post! My name is Manodeep, and I’m thrilled to embark on this journey of sharing my experiences, insights, and adventures as an aspiring Machine Learning (ML) engineer.

Who Am I?

To kick things off, let me introduce myself. I’m a curious soul with a passion for all things tech and AI. Currently, I’m diving deep into the fascinating world of Machine Learning and AI, exploring how these technologies can shape the future.

My Journey So Far

I’ve been on an exciting path of learning and discovery. Here are a few highlights:

  • Education: I’m a pre-final year student at KIIT University in Bhubaneshwar, India, pursuing a B.Tech in Electronics and Computer Science with a CGPA of 9.1.

  • Experience:

    1. I’m currently doing a summer deep learning internship at IIT Bhilai, where I’m working on a cancer prediction web app. This project combines the power of image classification using YOLOv8n and symptom-based prediction using a Retrieval-Augmented Generation (RAG) model with Mistral 7B. The app uses Gradio for an intuitive user interface.
    2. I also work as a member of RespAI Lab, focusing on LLM reasoning and LLM unlearning. Currently, I am contributing to a research paper exploring methods to improve reasoning capabilities and selective memory unlearning in LLMs.

Projects

Here are some of the exciting projects I’ve worked on:

  • RAG Chatbot for College: Developed a chatbot using Gemini 1.5 Flash, Huggingface, and LangChain for intelligent querying of PDFs, CSVs, and TXT files with customizable LLMs and FAISS for efficient vector storage.

  • Skin Cancer Classification: Created a skin lesion classification system using TensorFlow and Keras, implementing transfer learning models like MobileNet, VGG19, ResNet50, DenseNet121, and XceptionNet. Experimented with Vision Transformers (ViT), including a token-learning variant.

  • AkamaiCare Chatbot: Built a chatbot for hospital operations leveraging Gemini Flash 1.5 API, LlamaIndex, and Streamlit to handle natural language queries related to patient and hospital databases.

Skills Snapshot

  • Languages: Python, Java, C, SQL, HTML/CSS
  • Frameworks: Gradio, Flask, LangChain, Streamlit, Hugging Face, OpenCV, LlamaIndex
  • Libraries: Pandas, NumPy, Matplotlib, TensorFlow, Keras, PyTorch
  • Tools: Git, GitHub, Kaggle, Jupyter, Colab, VS Code, Linux, Roboflow

Why This Blog?

This blog is my way of sharing my learning journey, challenges, and breakthroughs. Whether it’s a cool project, a new technology, or just a nugget of wisdom I’ve picked up along the way, you’ll find it here.

Let’s Connect!

I believe that learning is a collaborative process. I’d love to hear from you! Whether you have questions, suggestions, or just want to say hi, feel free to reach out.

Thank you for joining me on this journey. Stay tuned for more updates, and let’s make the future smarter and brighter with AI!

Best regards,
Manodeep