Neural Networks and Deep Learning

Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization

Deep Learning Specialization

Machine Learning

Convolutional Neural Networks

Sequence Models

Structuring Machine Learning Projects

Bachelor of Engineering - BE

Machine Learning Engineer

1) Lead 4 member team of ML-engineers to develop CliffTs, an Unsupervised Time Series Anomaly Detection and Forecasting Framework.
2) Developed CliffOddity, a Community Outlier Detection System which detects irregular or outlier metrics out of a pool of metrics.

Machine Learning Lead

1) Developed Greendeck's Product Matching and Classification Functionality.
2) Created a multimodal fusion mechanism by effectively coupling image and text into a single modality using Corrnet and CLIP model. This resulted in an increase of 10% in matching accuracy over baseline.
3) Proposed a Novel mechanism to train a contrastive-based loss (Multi Similarity and Proxy Anchor Loss) with Adaptive Thresholds (depending on the hierarchy of classes in the dataset).
4) Created a model amalgamating Multimodal Fusion, DeepRank Model, Hierarchical Multi Similarity, and Proxy Anchor Loss. As a result, achieved 90% matching accuracy when evaluated by a human validator on the client's data.

Machine Learning Intern

Machine Learning Enginerr

Know about Keshav Gupta's work experience, education, bootcamps, certifications, and much more. Keshav Gupta