Denys Katerenchuk

Working on large language models for the past few years has taught me one crucial lesson: general-purpose models, no matter how capable, hit a wall when they encounter specialized domains like banking. At Kasisto, I’ve been driving development of KAI-GPT v4, and the journey has reinforced why domain expertise matters more than raw scale.

Working on large language models for the past few years has taught me one crucial lesson: general-purpose models, no matter how capable, hit a wall when they encounter specialized domains like banking. At Kasisto, I’ve been driving development of KAI-GPT v4, and the journey has reinforced why domain expertise matters more than raw scale.

Imagine a doctor says “The patient likely has pneumonia” versus “The patient has pneumonia.” That single word changes everything. The first statement hedges the diagnosis, expressing uncertainty that could affect treatment decisions. The second makes a definitive claim.

Imagine a doctor says “The patient likely has pneumonia” versus “The patient has pneumonia.” That single word changes everything. The first statement hedges the diagnosis, expressing uncertainty that could affect treatment decisions. The second makes a definitive claim.

Advanced algorithms improve our lives in subtle, but consistent ways. Google search results seldom make us go to the second page and Netflix magically recommends an interesting movie. These are outcomes of multiple trials and errors to find the best recommendation algorithm. How do these companies determine which algorithm is the best? Evaluating different ranking algorithms is not a trivial task. This is why we developed the Rank Discounter Cumulative Gain (RankDCG) evaluation measure.

Advanced algorithms improve our lives in subtle, but consistent ways. Google search results seldom make us go to the second page and Netflix magically recommends an interesting movie. These are outcomes of multiple trials and errors to find the best recommendation algorithm. How do these companies determine which algorithm is the best? Evaluating different ranking algorithms is not a trivial task. This is why we developed the Rank Discounter Cumulative Gain (RankDCG) evaluation measure.

For over ten years, my main work tool has been a laptop. For most tasks — whether it is software development or research — laptops deliver good performance with the advantage of working from anywhere. More demanding work gets done on remote servers or cloud computing platforms. In my lab, I set up a Sun Grid Engine cluster of six servers that brings together the power of 36 Xeon CPUs. This system was used to build Automatic Speech Recognition (ASR) models (which took over a month to train) and to run many other extensive experiments. My laptop simply couldn’t compete with it. However, in recent years with the development of deep learning, GPUs have become incredibly powerful tools in research. For example, when I needed to run the same ASR training, a single GPU reduced training time from over 30 days to around 6.

For over ten years, my main work tool has been a laptop. For most tasks — whether it is software development or research — laptops deliver good performance with the advantage of working from anywhere. More demanding work gets done on remote servers or cloud computing platforms. In my lab, I set up a Sun Grid Engine cluster of six servers that brings together the power of 36 Xeon CPUs. This system was used to build Automatic Speech Recognition (ASR) models (which took over a month to train) and to run many other extensive experiments. My laptop simply couldn’t compete with it. However, in recent years with the development of deep learning, GPUs have become incredibly powerful tools in research. For example, when I needed to run the same ASR training, a single GPU reduced training time from over 30 days to around 6.

Introduction

Can public opinion predict stock market movements? With recent advancements in natural language processing (NLP), this question has gained significant attention in both academic and financial circles. As an NLP scientist, I decided to investigate this relationship and satisfy my own curiosity about the intersection of social media sentiment and financial markets.

Introduction

Can public opinion predict stock market movements? With recent advancements in natural language processing (NLP), this question has gained significant attention in both academic and financial circles. As an NLP scientist, I decided to investigate this relationship and satisfy my own curiosity about the intersection of social media sentiment and financial markets.

Abstract

This study presents a comprehensive text analysis of the pickup dating community through the lens of social network analysis, natural language processing, and community psychology. Working alongside Anders Wallace, an Anthropology Ph.D. student, we analyzed forum data from fastseduction.com to understand community structure, social hierarchies, and linguistic patterns. Our findings reveal complex social dynamics, educational diversity, and hierarchical structures that challenge common stereotypes about this community.

Abstract

This study presents a comprehensive text analysis of the pickup dating community through the lens of social network analysis, natural language processing, and community psychology. Working alongside Anders Wallace, an Anthropology Ph.D. student, we analyzed forum data from fastseduction.com to understand community structure, social hierarchies, and linguistic patterns. Our findings reveal complex social dynamics, educational diversity, and hierarchical structures that challenge common stereotypes about this community.

[Note: Unfortunately, most images were lost due to changes in carto.com service.]

[Note: Unfortunately, most images were lost due to changes in carto.com service.]

Introduction

In this blog post, we will explore various data visualization techniques using datasets from New York City. Our primary focus will be on analytics related to Broadway shows and how we can extract meaningful insights from the data.

Introduction

In this blog post, we will explore various data visualization techniques using datasets from New York City. Our primary focus will be on analytics related to Broadway shows and how we can extract meaningful insights from the data.

This year my colleagues from SpeechLab and I were invited to Interspeech2014 conference to present our work. The conference was held in Singapore, a great city with multilingual culture. The city was the perfect choice for an NLP and speech conference. I was impressed by how well everything was organized from the opening ceremony to the very end. It made the 24-hour flight from NYC to Singapore well worth it. The best part of the conference was the whole slew of interesting research papers in various areas of NLP.

This year my colleagues from SpeechLab and I were invited to Interspeech2014 conference to present our work. The conference was held in Singapore, a great city with multilingual culture. The city was the perfect choice for an NLP and speech conference. I was impressed by how well everything was organized from the opening ceremony to the very end. It made the 24-hour flight from NYC to Singapore well worth it. The best part of the conference was the whole slew of interesting research papers in various areas of NLP.