Dr. Egor Kraev

Creating structured objects: the most impactful application for GenAI in production

When people think of usecases for Large Language Models (LLMs), they usually focus on producing free form text, or maybe code. Yet by far the most useful application of LLMs in my experience is their ability, with the right scaffolding, to take squishy data, such as natural language or images, and use it to generate highly complex objects with predefined structure. Recent examples from my practice include producing complex chart configs from plain images, and precisely structured semantic layer queries from natural language questions by the user. Having the model return an exact, validated Pydantic object, maybe several levels deep, allows for a degree of control and reliability that is not achievable with naive approaches such as direct SQL generation - so much so that prompt tuning or the exact choice of model hardly even matter. I’ll explain how to do that reliably and easily with open source tools, using a combination of structured outputs and an agentic feedback loop.

Dr. Egor Kraev has been applying machine learning to real-world problems since the last millennium, including economic and human development data analysis for nonprofits in the US, the UK, and Ghana, and 10 years as a quant, IT solutions architect, and occasional trader at UBS then Deutsche Bank. Egor's last role was Head of AI Wise, where over 5 years he brought the power of ML to bear in a variety of domains, from fraud detection to trading algorithms, and causal inference for A/B testing and marketing, as well as multiple GenAI projects and workshops across the company. During that time, Egor took the Wise Data Science team from an idea to a well-structured team of over 30 people. Egor is the lead author of several open-source data science packages such as wise-pizza (finding interesting segments in multidimensional data), causaltune (AutoML for causal inference estimators) and motleycrew.ai (mix and match different agentic frameworks, with some twists he hasn't seen in any other framework). In April 2025, Egor has left Wise to co-found motley.ai, which aims to take number-driven storytelling in firms to the next level of usability and power.