ML Internship

  | #Deep Learning#Machine Learning#Recommendation Engine

Collaborated Projects

Adversarial De-biasing for Semantic Text Classification

Developed a Text Classification application where the result are fair and unbiased from gender or racial data in input text.

Technologies

python PyTorch sk_learn pandas

Random Attack Pipeline Revamp

Debug and Upgrade the ML model evaluation and stress analysis pipeline, from the aspects of MLOps.

Technologies

python C C++ bash jenkins docker

Recommendation Personalized Re-ranking Application

Rank the suggestion generated by the inherent or external recommendation model.

Technologies

python h2o_wave numpy pandas sk_learn keras

Recommendation Pre-model Analysis Application

Analysis compatibility and complexity of a dataset from the aspect of recommendation model training.

Technologies

python h2o_wave numpy pandas sk_learn

RecSys Application

Application for train and evaluate recommendation engines & benchmark those models with advance fine-tuning.

Technologies

python h2o_wave numpy pandas sk_learn keras tensorflow PyTorch

Recommendation Post-model Analysis Application

Comprehensive Drift and Fairness analysis on recommendation models.

Technologies

python h2o_wave numpy pandas sk_learn

Alternative Credit Risk Scoring Application

Personal Credit risk scoring based on pattern recognition in Telco data.

Technologies

python h2o_wave numpy pandas sk_learn keras tf

AI Maturity Index

Company AI use and in-house capacity analysis and ranking application based on their job description data.

Technologies

python h2o_wave numpy pandas matplotlib 3djs selenium beautifulsoup sk_learn

Key Achievements

  • Developed a sentiment analysis model with an accuracy improvement of 15%
  • Led a team of data scientists in building a recommendation system that increased user engagement by 20%
  • Implemented a production-ready chatbot using state-of-the-art NLP techniques, reducing customer support response time by 30%