Aatish Kayyath
Development
New York, United States
Skills
Data Science
About
AATISH KAYYATH's skills align with IT R&D Professionals (Information and Communication Technology). AATISH also has skills associated with Programmers (Information and Communication Technology). AATISH KAYYATH appears to be a low-to-mid level candidate, with 4 years of experience.
Work Experience
Data Science Intern
WORLD WIDE TECHNOLOGY
May 2023 - August 2023
- Developed a 3D image segmentation model utilizing NVIDIA's MONAI framework for precise pancreas and tumor detection in CT scans, employing UNET, SEGRESNET, and VNET architectures, resulting in a 15% improvement in Dice coefficient accuracy. Optimized for parallel processing efficiency on NVIDIA A100 MIG, and deployed via Streamlit Engineered a Langchain and BERT-enhanced chatbot, achieving streamlined HR inquiry processing and a 10x increase in simultaneous inquiry handling
Senior Data Scientist
EVALUESERVE
July 2021 - August 2022
- HSBC Bank: Designed an OCR model to expedite the KYC verification step in the loan sanctioning process, increasing throughput by over 300%. Leveraged CNN and RNN for OCR and classification tasks and hosted the app on GCP McDonalds: Implemented a generalized price forecasting dashboard using ARIMA and LASSO regression, resulting in savings of over $2M per quarter
Data Scientist
IMPACT ANALYTICS
January 2019 - July 2021
- Puma: Implemented a Named Entity Recognition (NER) system utilizing FashionBERT to parse and extract detailed product attributes from textual descriptions, achieving a 250-fold enhancement in processing efficiency. Belk: Developed dynamic assortment planning tool leveraging KMeans, AP Clustering, Lasso regression, and optimizers like PuLP. This tool, handling vast amounts of real-time data upwards of 2TB hosted on Snowflake, minimized inventory over-buys and avoided stockouts, boosting the profit margin by 3.2%. PVH (Tommy Hilfiger, Calvin Klein): Standardized the assortment tool and made a plug-and-play interface which was then utilized by 12+ clients BJs: Conceptualized an A/B testing tool to automate the selection process and help organizations run seamless experiments which was then used to test over 100+ live campaigns
Data Science Intern
VASPP TECHNOLOGIES
June 2018 - August 2018
- Built various key identifier models and What-If Analysis model and deployed it using ML Studio on Azure
Education
STEVENS INSTITUTE OF TECHNOLOGY
MS in Computer Science
July 2022 - May 2024
