Tatiana Verbitskaya

The topic of fine-tuning large language models arises when it becomes apparent that general-purpose LLMs do not always behave as expected in specific tasks. A general model may be effective at supporting dialogue or answering broad questions, but when dealing with narrow topics, specialized terminology, or fixed-response formats, its behavior

Artificial Intelligence has been an integral part of modern business for quite some time. The emergence of large language models marked a qualitatively new stage of powerful systems capable of understanding, generating, and summarizing human language on an unprecedented scale. Language is the foundation of any business activity: from customer

LLM is a large language model (large language model meaning), a type of language model trained on massive amounts of text to predict and generate natural language. These models are powered by core assistants like ChatGPT, Gemini, Claude, and Perplexity. They provide summarization, translation, reasoning, and conversational interfaces that transform

The modern robotics landscape is based on specialized software that enables machines to see the world, make decisions, and safely interact with complex environments. As robots move from controlled laboratory settings to factories, warehouses, farms, medical facilities, and public spaces, the software infrastructure behind their operation becomes increasingly complex and

In the modern world, where pupils and students write thousands of essays and short answers, a significant problem arises: how to check them quickly and objectively? Manual checking has always been slow and expensive, and, most importantly, it involves a subjective element. The scores for the exact text can vary

Deforestation Monitoring for AI-powered Tagging utilizes a decadal set of satellite-derived climate data and related information to detect habitat loss over multiple years. It combines programmatic access, standard frameworks, and native Python tools to enable teams to quickly discover and process information. This approach reduces noise and seasonality by utilizing