Skip to content
View samy19980109's full-sized avatar
🏠
Working from home
🏠
Working from home

Sponsoring

@thedotmack

Block or report samy19980109

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. 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
samy19980109/README.md

Call me Samarth or Sam

I'm a senior Machine Learning Engineer designing and developing creative software/data solutions to enhance business productivity, gain insights & generate revenue.

  • πŸ”­ Currently working as a Machine Learning Engineer - Senior Specialist at Vanguard Investments Canada, developing Gen AI solutions for Financial Advisors.
  • 🌱 Proficient in Machine Learning techniques, data analytics, warehousing, and workflow orchestration (Apache Airflow, Kubeflow).
  • πŸ’‘ Experienced in developing advanced CNN fingerprinting solutions (eBay, reducing fraud by 95%) and XGBoost models for churn prediction (Corus Entertainment, 90% accuracy).
  • πŸ’¬ Ask me about anything related to ML pipelines, cloud platforms (AWS, GCP, Azure), or full-stack development.
  • πŸ“« How to reach me: E-mail or Linkedin.
  • ⚑ Fun fact: Winner of Toronto's largest hackathon, UofTHacks, in 2019.

πŸš€ Languages and Tools:

My Skills


Work Experience Highlights

  • Machine Learning Engineer - Senior Specialist @ Vanguard Investments Canada (01/2026 - Present): Developing Gen AI solutions.
  • Software Engineer - AI & ML @ eBay: Engineered a CNN fingerprinting solution to combat return fraud, reducing it by 95%. Developed pipelines for Computer Vision models.
  • Machine Learning Engineer @ Corus Entertainment: Implemented AWS Rekognition and GCP for video content analysis and an XGBoost model for customer churn prediction.

Education

  • B.Sc.(Hons.) Mathematics, Statistics and CS @ University of Toronto

Personal Projects

  • Real-Time Speculative Decoding Visualization (01/2026): A full-stack system achieving 3.4x inference speedup for LLMs using speculative decoding.
  • High-Performance LLM Evaluation Framework (12/2025): Built a four-tier evaluation framework with a logit drift guard for numerical regression detection on custom silicon.
  • Multi-Agent Retrieval-Augmented Generation (RAG) system (12/2024): Engineered an efficient AI application using Langflow, OpenAI API, and Streamlit.
  • Fine Tuning LLM with QLoRa (06/2023): Fine-tuned a 13B parameter LLM (Llama) using 4-bit quantization.

Pinned Loading

  1. FIT-TCX-Chatbot FIT-TCX-Chatbot Public

    Chatbot which accepts your FIT and TCX Files from Strava, Garmin, Amazfit and more and does AI analysis and heart rate analysis on

    Python

  2. github_trending_repos_to_whatsapp github_trending_repos_to_whatsapp Public

    Send Daily Github Trending Repos on Whatsapp

    TypeScript

  3. AeroEval-Eval-Framework AeroEval-Eval-Framework Public

    High-performance LLM evaluation framework with four-tier scoring, designed for CI/CD integration to prevent model regressions.

    Python

  4. Visualize-Speculative-Decoding Visualize-Speculative-Decoding Public

    Visualize Speculative Decoding

    Python

  5. OCR_with_Nanonets_3B_on_Apple_Silicon_using_MLX OCR_with_Nanonets_3B_on_Apple_Silicon_using_MLX Public

    OCR with Nanonets 3B on Apple Silicon using MLX. The model is quantized to 4 bit to fit on my Macbook

    Python 1

  6. Mixture-of-Experts-Visualizer Mixture-of-Experts-Visualizer Public

    Visualize Mixture-of-Experts (MoE) when making a inference call on Apple Silicon using MLX

    TypeScript