Tashko Pavlov

Data Еngineer, Kin + Carta Skopje

Tashko Pavlov is a Data Scientist at Kin + Carta. He graduated at the Faculty of Computer Science and Engineering in Skopje with a bachelor’s degree in Software Engineering. In his career, Tashko has worked on many ML projects and has done several researches in the domain of NLP. His utmost favorite one is the Macedonizer, where together with a research group at the Faculty of Computer Science and Engineering he trained a transformer model for the Macedonian language for the first time. In his free time he enjoys hiking, mountain biking and football on Sundays.

___________________ 🎙️ Chat with your own data by using LLaMA2 and LangChain 🎙️ In today's world data is everywhere and generated every second - in structured, semi structured but mostly unstructured format. Yet, often data experts face the challenge of too much data to analyze. Along with that, we know how sensitive everyone is about their data. So how can data scientists process data on prem without the hidden background of some conversational models? The short answer is - with a system that can answer questions about data in a human-like way. In his talk, Tashko will dive deeper into how we can use open source models and libraries to develop chatbots for specific use cases. He will address queries like: ✔️ How to build your own custom chatbot? ✔️ Understand the benefits of using open source libraries? ✔️ What are the pros and cons of open vs closed-source LLMs?