Professional
Niels Kasch, Ph.D.
I was the co-founder of Miner & Kasch, a leading AI and Data Science firm. Until ...
NVIDIA GTC 2025: Agentic AI at the Edge: Real-Time Sentiment Analysis for Engaging Customer Interaction
Discover an AI agent that reimagines customer interactions, turning physical spaces into engaging and personalized experiences. Using on-device Vision Transformers (ViTs) on NVIDIA Jetson, this solution performs real-time sentiment analysis to gauge audience reactions while providing relevant product contents and ensuring rapid responsiveness. It processes sensitive data locally, sending only anonymized metadata to the cloud via Kafka streams, and employs hybrid cloud reinforcement learning and generative AI to tailor advertisement content dynamically. Leveraging NVIDIA DeepStream for real-time video analytics and a closed-loop feedback mechanism, the system continuously improves interactions with customers to maximize their experiences and improve business ROI.
Transform your NLP Skills: Using BERT (and Transformers) in Real Life (ODSC 2020)
This workshop teaches you the use of transformer neural networks and their incarnations (BERT, RoBERTa, GPT-2) for solving real-world natural language use cases. NLP has advanced tremendously over the last few years and BERT is at the forefront of this success having achieved state-of-the-art results on 11 different NLP tasks. For businesses, BERT has unlocked new NLP use cases that have been previously unattainable.
Highlights from the Maryland Data Science Conference: Deep Learning on Imagery and Text
Read the post on Domino's blog.
Getting value from data using modern approaches and data platforms
Corelogic Epiq 2018 (invited talk)
See below for a collections of older things:
Financial Compliance: New Frontiers with Data Science
This blog explains new ways to help financial institutions address compliance. Since the world of regulations has changed, the existing compliance and governance technology and infrastructure platforms fail to meet new business requirements, often covering only archival or basic analytics. By implementing a data lake, advanced data science algorithms, and user interfaces that build analyst feedback into predictive models, compliance groups can reach a new level of operational excellence.
Podcast - "Discussing natural language processing and churn analysis”
A widely published podcast on the technology and data science behind real-life natural language processing needs and a deep dive on financial churn prediction. (Link to podcast)
3 Key Capabilities Necessary for Text Analytics & Natural Language Processing in the Era of Big Data
A series of blogs to introduce text analytics and showing its applicability in a business context. The blog illustrates several industry use-cases where text analytics and NLP are necessary tools to address real world business needs and describes the technology to address these use-cases in a big data environment. (Read the series)
Text analytics and natural language processing in the era of big data
A series of blogs to introduce text analytics and showing its applicability in a business context. The blog illustrates several industry use-cases where text analytics and NLP are necessary tools to address real world business needs and describes the technology to address these use-cases in a big data environment. (Read the series)
Churn Prediction in Retail Finance and Asset Management (Part 2)
A series of blogs detailing the data science challenges involved in building predictive and explanatory models for churn prediction. The series gives an in depth view of modeling including data assets, analytics algorithms, and resulting customer applications. (Read the series)
Churn Prediction in Retail Finance and Asset Management (Part 1)
A series of blogs detailing the data science challenges involved in building predictive and explanatory models for churn prediction. The series gives an in depth view of modeling including data assets, analytics algorithms, and resulting customer applications. (Read the series)