Machine Learning Engineer | MS CS (ML) @ Santa Clara University
Recommender Systems • Retrieval • Applied ML Systems
I’m a Machine Learning Engineer with experience building production ML systems and applied research prototypes, spanning conversational AI, recommender systems, and vision-language models.
Previously worked as a Software Engineer at Haptik (Jio), contributing to large-scale conversational AI platforms used by 10k+ businesses, and as an ML Intern at a stealth startup building RAG-based AI agents for consumer enterprises.
Currently pursuing an MS in Computer Science (ML specialization) at Santa Clara University.
- Recommender systems (ranking, retrieval, CTR prediction)
- Open-set and zero-shot ML (CLIP, vision-language models)
- Representation learning & evaluation
- Applied ML systems: data pipelines, metrics, scalability
- NLP & conversational AI
- Built a zero-shot image retrieval system using CLIP ViT-B/32
- Supports image-to-image and text-to-image retrieval on unseen classes
- Achieved 98.1% Recall, outperforming fine-tuned CNN baselines with 90% less compute
- Tech: PyTorch, OpenCLIP, FAISS, t-SNE
👉 Repository coming soon
- Trained ML models on millions of ad impressions
- Implemented GBDT, Random Forests, and deep models (DCNv2)
- Improved AUC / LogLoss over baseline feature sets
- Focused on feature engineering for users, ads, and interactions
👉 Repository coming soon
Languages: Python, SQL
ML: PyTorch, TensorFlow, scikit-learn, NumPy, Pandas
Data & Infra: AWS, Airflow, MongoDB, Linux
Visualization: Matplotlib, Seaborn, Power BI, Tableau
- LinkedIn: https://linkedin.com/in/harshvardhan-garude
- Email: hgarude@scu.edu