kchatr / calorme Goto Github PK

View Code? Open in Web Editor NEW

CalorMe: A mobile and web app that retrieves nutritional information based on a captured or uploaded image of the desired food using a CNN.

Kotlin 0.92% Swift 3.00% Objective-C 0.28% Dart 67.21% HTML 28.59%

calorme's Introduction

Welcome to my Github :D

Intro

I'm Kaushik - a 20-year-old AI developer and Data Science enthusiast. I'm especially passionate about responsible AI development, with ethical alignment and interpretability my primary areas of interest. I have significant experience working with retrieval-augmented generative (RAG) models, as well as autoregressive and classification algorithms.

🔭 I recently interned as a Data Scientist at ProCogia, where I developed autoregressive models for real-time data, as well as benchmarking LLms and implementing guardrails. Previously, I developed, evaluated, and fine-tuned retriever and generative models in RAG systems and deployed LLMs to REST API endpoints.
🌱 I’m currently learning more about 3D model generation in the context of spatial computing, as well as developing an interest in representation learning and multimodal techniques.
💬 Ask me about anything related to generative AI, machine learning, or philosophy :)
📫 How to reach me: My LinkedIn (https://linkedin.com/in/kchatr). You can check out my articles at https://medium.com/@kchatr

🧰 Toolkit