Senior Software Engineer @ Google. Focus in Machine Learning and Recommendation Systems. Find out more about me at my website below
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Data Scientist
June 25, 2026 – Present
Amazon-Video-Game-Recommender-System-PyTorch-Transformer-Model
May 27, 2026 – Present
Sequential recommendation on Amazon Video Games, built progressively from a bag-of-items baseline to a full self-attention transformer — measuring what attention and positional embeddings each contribute. Streamlit demo included.
View ProjectPokemon-TCG-Price-Tracker
May 24, 2026 – Present
script to track prices of watch-listed Pokemon TCG cards and alert user based on certain rules
View ProjectGame-Recommender-System-PyTorch-TwoTower-Model
April 22, 2026 – Present
Two-stage recommender on Steam — two-tower retrieval feeds a Wide & Deep ranker that rescores candidates with user×item cross-features a dot product can't capture. +16% NDCG@10 over retrieval-only. Streamlit demo lets users view retrieval-only recs or compare against ranker-reranked recs.
View ProjectBook-Recommender-System-PyTorch-TwoTower-Model
April 8, 2026 – Present
Two-tower retrieval model on Goodreads — full softmax over the catalog, behavior-partitioned user pools, and dedicated author/shelf-affinity towers. 3.4× Hit@10 over MSE baseline. Deployed to Streamlit.
View ProjectMovie-Recommender-System-PyTorch-Transformer
June 4, 2025 – June 4, 2025
Create a movie recommender system by building a Transformer neural network in PyTorch
View ProjectMovie-Recommender-System-PyTorch-TwoTower-Model
May 28, 2025 – Present
Two-tower retrieval model on MovieLens 32M — full softmax over the catalog, behavior-partitioned user pools, and content-aware item tower with genre and genome tag features. 8.7× MRR over MSE baseline. Deployed to Streamlit.
View Projectcu2rec
November 8, 2018 – April 15, 2019
CUDA Implementation of Parallel Matrix Factorization Algorithm for Recommender Systems
View Projectnews_crawler_classifier
September 25, 2017 – March 15, 2018
A Python program to crawl RSS feeds and use sklearn classification algorithms on the results
View ProjectCultural Fit Analysis
The candidate's project portfolio shows a strong inclination towards personal learning and development, with a focus on cutting-edge machine learning techniques. The projects are diverse within the realm of data science, particularly recommender systems, which aligns well with an innovative and research-oriented culture. However, the lack of team-based or professional projects makes it difficult to fully assess collaboration and broader cultural fit.
Soft Skills & Operational Fit
Insufficient data to assess soft skills or operational fit. The candidate's project descriptions are clear and concise, but no direct communication or collaboration assessments are available.