Kuzu, Knowledge Graphs, and the AI Revolution with Prashanth Rao

In this episode of the Hedgineer Podcast, host Michael Watson is joined by Prashanth Rao, AI Engineer at Kuzu, for a deep dive into the world of embedded graph databases and their pivotal role in the age of AI. Building on the previous week's discussion of the columnar database DuckDB, this conversation explores Kuzu, a parallel concept focused on creating high-performance, embedded graph databases.

Michael and Prashanth explore why Kuzu's unique architecture—an embedded, columnar, and strictly-typed system—is delivering incredible speed and scalability for complex analytical queries. They discuss the resurgence of interest in knowledge graphs, driven by the need to impose structure on data for modern AI and LLM workflows. Prashanth explains how LLMs are revolutionizing both the upstream construction of graphs from unstructured data and the downstream querying of these graphs by translating natural language into the Cypher query language.

Tune in to learn about:

• What Kuzu Is: An in-process, embedded graph database designed from the ground up for query speed and scalability, blending the benefits of columnar processing with the property graph data model.
• AI and Graph Synergy: How large language models (LLMs) assist in both building knowledge graphs through structured output extraction and accessing them via natural language to Cypher translation.
• Modern Data Stack Integration: The role of Kuzu as a powerful secondary semantic layer that interoperates seamlessly with primary data stores in data lakes (like Parquet files on S3), DuckDB, and Postgres.
• Programming with LLMs using DSPy: A detailed look into using DSPy to create structured, modular, and optimizable prompts for more reliable LLM applications, bridging the gap between deterministic code and fully agentic workflows.
• The Enterprise Knowledge Graph: A discussion on how a knowledge graph can become the single source of truth for understanding complex organizational workflows, data lineage, and interdependencies across an entire enterprise.
• The Future of Kuzu: A preview of what's next for the Kuzu project, including new graph algorithm capabilities and enhanced concurrency for reads and writes.

Whether you're a data engineer, AI practitioner, or a leader in the asset management space, this episode provides a masterclass on leveraging modern database technology to build powerful, scalable, and intelligent applications.

Hedgineer.io

Follow The Hedgineer Podcast:

• YouTube: (https://www.youtube.com/@hedgineer)
• LinkedIn: (https://www.linkedin.com/company/90976838)
• Twitter: (https://x.com/hedgineering)
• Instagram: (https://www.instagram.com/hedgineer/)

Hosted on Acast. See acast.com/privacy for more information.

In this episode of The Hedgineer Podcast, host Michael Watson is joined by special guest Rusty Conover, the world's most prolific DuckDB extension builder, for a masterclass on building the next generation of real-time, large-scale data systems.

Rusty, who has an extensive career in data engineering, including at multi-manager hedge funds, pulls back the curtain on what makes DuckDB so revolutionary for developers and data engineers. They explore how its blazingly fast, in-process, C++-based architecture is challenging the big data status quo. The conversation provides a deep dive into the powerful ecosystem growing around DuckDB, from the Apache Arrow columnar format to the evolving landscape of open table formats like Iceberg, Delta Lake, and the new DuckLake.

Join them for a detailed discussion on the nitty-gritty of modern data infrastructure, whether you're building enterprise data platforms or looking for the most efficient tools for your analytics workload.

In this episode, you will learn about:

The DuckDB Revolution: What makes this "blazingly fast" in-process database a game-changer that can simplify and replace entire ETL stacks.

A Tour of DuckDB Extensions: A look inside some of the 15 extensions Rusty has built, from Airport for integrating with Apache Arrow, to Crypto, ShellFS, and TextPlot.

Diving into Apache Arrow: An explanation of the columnar in-memory data format, zero-copy operations, and the Arrow Flight RPC mechanism for efficiently moving data.

The Battle of Open Table Formats: A comparison of Iceberg, Delta Lake, and the new database-centric approach of DuckLake.

DuckDB vs. The World: How DuckDB stacks up against KDB for financial data, ClickHouse for analytics, and its role alongside large-scale compute engines like Apache Spark.

Parquet Deep Dive: The key differences between Parquet V1 and V2 and the importance of modern compression strategies and encodings.

The Future of DuckDB: A sneak peek at powerful upcoming features like time travel and the MERGE INTO statement for simplifying change data capture (CDC) pipelines.

Hosted by Michael Watson, The Hedgineer Podcast dives into AI technology and data in the hedge fund, asset management, and prop trading space.

Follow The Hedgineer Podcast:

YouTube: (https://www.youtube.com/@hedgineer)

LinkedIn: (https://www.linkedin.com/company/90976838)

Twitter: (https://x.com/hedgineering)

Instagram: (https://www.instagram.com/hedgineer/)

Don't forget to like, subscribe, and hit the notification bell to stay updated on our latest episodes!

Hedgineer.io

Hosted on Acast. See acast.com/privacy for more information.

Welcome back to Season 2 of The Hedgineer Podcast, where host Michael Watson talks data, AI, and technology in the hedge fund and asset management space.

In this episode, Michael sits down with Didier Lopes, the founder of OpenBB, for a conversation that started five years ago during the meme stock boom. Michael reached out to Didier in 2021 after he shared his "Gamestonk Terminal" on r/wallstreetbets. Michael reached out to Didier if he was interested in a job on his team at Citadel but Didier had other ideas. That initial LinkedIn message from Michael ended up in the pitch deck that helped Didier raise nearly $9 million for his company and the rest is history.

Didier recounts his journey from being a software engineer who automated his personal investment research to becoming the creator of a massively popular open-source project. The Gamestonk Terminal went viral on WallStreetBets and Hacker News, gaining 4,000 GitHub stars in less than 24 hours by cleverly aggregating the free API tiers from hundreds of different data vendor and has evolved into what OpenBB is today.

Join us as we dive into the evolution of a viral tool into a sophisticated enterprise platform and explore the disruptive forces reshaping the financial data landscape.

In this episode, you will learn about:

• The Origin Story: How a canceled flight during the COVID-19 pandemic led to the creation of the Gamestonk Terminal.
• From CLI to Enterprise: The strategic pivot from a command-line tool for retail investors to a customizable, SDK-first workspace for financial firms (and then possibly back to CLI with agents).
• The Data Vendor's Dilemma: A fascinating discussion on the conflict facing companies like FactSet and Capital IQ, whose UI and terminal businesses are threatened by the rise of AI and direct data access through protocols like MCP (Model Context Protocol).
• The Future of Workflows: Michael explains how he uses AI agents like Claude Code combined with CLI tools to build powerful, autonomous systems for everything from risk management to analyzing entire enterprise data warehouses.
• The Autonomous Firm: A look into a future where AI enables multi-billion dollar asset managers to operate with fewer than ten people.
• The Rise of "Acqui-hires": Why the next wave of M&A might value a team's ability to leverage AI far more than the IP they've built.

This is episode is for anyone at the intersection of finance, data science, and artificial intelligence.

Hedgineer.io

Follow The Hedgineer Podcast:

• YouTube: (https://www.youtube.com/@hedgineer)
• LinkedIn: (https://www.linkedin.com/company/90976838)
• Twitter: (https://x.com/hedgineering)
• Instagram: (https://www.instagram.com/hedgineer/)

Hosted on Acast. See acast.com/privacy for more information.