Skip to content
View daniel-mehta's full-sized avatar
😼
😼
3 followers · 6 following

Achievements

Achievement: QuickdrawAchievement: Pull Shark

Achievements

Achievement: QuickdrawAchievement: Pull Shark

Block or report daniel-mehta

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
daniel-mehta/README.md

👋 Hey, I'm Daniel Mehta

🧠 About Me

Wilfrid Laurier CS & UX grad · Humber AI & ML postgrad graduate
Former Data Analyst and Database Intern at the Ministry of Education and DPCDSB
Some startup experience at Meter, a VC-backed SF company funded by Sam Altman, Sequoia Capital, and Stripe’s founders

🔍 Currently Exploring

  • Local LLM Deployment - building privacy-friendly, offline AI solutions with Ollama and LangChain
  • Multimodal AI & VLMs - combining vision and language for practical tools like personal stylists and OCR pipelines
  • Advanced Model Optimization - LoRA fine-tuning, quantization, and efficient transformer training
  • Business AI Applications - PySpark-based customer segmentation and LRFS modeling
  • Sports Analytics - improving predictive accuracy for Fantasy Premier League points
  • Game Development - designing a low-poly survival horror in Unreal Engine 5 for Steam release

🛠 Tech Stack

Languages:
Python · Java · C# · R · Go · Julia · SQL

Machine Learning & AI:
Scikit-learn · TensorFlow · PyTorch · Hugging Face · LangChain · AutoML · AutoGen · Vision-Language Models (VLMs) · OpenCV · PySpark

Data & Analytics:
Pandas · NumPy · Matplotlib · Tableau

Cloud & Deployment:
AWS (EC2, S3, SageMaker) · Flask · FastAPI

Databases:
PostgreSQL · MySQL · MongoDB

📬 Reach Out

Gmail
LinkedIn

Pinned Loading

  1. CooperLM-354M CooperLM-354M Public

    A 354M parameter GPT-2 model trained on filtered Wikipedia, BookCorpus, and OpenWebText. Includes 4-bit quantized version for lightweight inference. Built as a toy project to explore end-to-end LLM…

    Jupyter Notebook 1

  2. FitCheck.AI FitCheck.AI Public

    👔 FitCheck.AI is your personal AI stylist. Upload outfits for savage critiques, auto-tag your wardrobe, and get smart recommendations - powered by Streamlit, LangChain, MongoDB, and VLMs like CLIP …

    Python 2 1

  3. Code-Reviewer Code-Reviewer Public

    A lightweight, LLM-powered tool for reviewing Python code using DeepSeek-Coder via Ollama. The backend was built in a day, with a Streamlit frontend added later for a clean, local code review exper…

    Python

  4. Traffic-Management-Q-Learning Traffic-Management-Q-Learning Public

    An artificial intelligence model for optimizing traffic flow and managing autonomous vehicles using Q-learning. Implements the BNART heuristic for efficient routing.

    Python

  5. AWS-Sentiment-Analysis-Project AWS-Sentiment-Analysis-Project Public

    Real-time sentiment analysis web app using AWS Comprehend, built with Flask and deployed on AWS EC2.

    Python

  6. Salary-Predictor Salary-Predictor Public

    Multi-language salary prediction pipeline using Julia (MLJ), Go, SQL, Python and Bash — trained on Kaggle data with a Go-powered web UI.

    Go 1