Victor Giannakouris

About Me

I am a final-year Ph.D. student in Computer Science at Cornell University, advised by Immanuel Trummer. My research focuses on building GenAI-native data systems that integrate Large Language Models (LLMs) to improve core database functionalities. I am also deeply interested in database internals, particularly query optimization across heterogeneous engines—including federated and learned query optimization. Throughout my academic and industry experience, I have built custom components for a variety of data systems, including Spark SQL, TileDB, Dremio, Presto/Trino, and MonetDB.

I am originally from Amarynthos, a coastal town in Evia, Greece.

Research Interests

• Large Language Models & Data Systems: Systems that harness Large Language Models to enhance and optimize data systems (GenAI for Data Systems) and design GenAI-native systems that integrate LLMs into their core functionality (Data Systems for GenAI).
• Query Optimization: Federeted Query Optimization, Learned Query Optimization, Query Engines, Internals

Education

• Ph.D. in Computer Science (Current)
  Cornell University, Ithaca, NY, USA
  Advisor: Immanuel Trummer

• M.Sc. in Computer Science (2024)
  Cornell University, Ithaca, NY, USA
  Advisor: Immanuel Trummer

• M.Sc. in Computer Science (2021)
  National and Kapodistrian University of Athens, Athens, Greece
  Advisor: Yannis Ioannidis

• B.Sc. in Informatics (2017)
  Ionian University, Corfu, Greece
  Advisor: Dimitrios Tsoumakos

Professional Experience

• Research Intern (Summer 2025)
  Gray Systems Lab, Microsoft Research, Mountain View, CA, USA

• Research Intern (Summer 2023)
  Data Systems Group, Microsoft Research, Redmond, WA, USA

• Research Intern (Summer 2022)
  Almaden Research Center, IBM Research, San Jose, CA, USA

• Senior Software Engineer (2019 - 2021)
  TileDB Inc., Boston, MA, USA

• Software Engineer (2018-2019)
  Unravel Data Systems, Palo Alto, CA, USA

• Data Engineer (April 2018- Oct. 2018)
  Aisera, Palo Alto, CA, USA

Projects

A small summary of my projects is provided below. For more details, refer to the publications section.

SwellDB: GenAI-Native Query Processing via On-the-Fly Table Generation

SwellDB is a GenAI-native data system that generates tables on-the-fly using LLMs.

λ-Tune: LLMs for Database System Tuning

A system that uses Large Language Models to tune database systems. It leverages LLMs to analyze the input workload (SQL queries) and generate full configuration scripts.

DBG-PT: Query Plan Debugging using Large Language Models

A system that leverages LLMs to resolve query plan regressions. It uses the LLMs in order to compare a regressed query plan with a previously efficient one and recommend workarounds.

A Sample Index for MonetDB

An index structure for MonetDB that accelerates the sampling operator (SAMPLE clause). It is based on random number generation and priority queues (min-heaps). The thesis can be found in this link.

News

• [June 2025] My proposal with title “SwellDB: GenAI-Native Query Processing via On-the-Fly Table Generation” was accepted at VLDB 2025 PhD Workshop!
• [June 2025] Our paper “Rethinking Federated Query Optimization: From Laptops to Data Warehouses” was accepted at CDMS at VLDB 2025!
• [May 2025] I am joining the Gray Systems Lab at Microsoft as a Research Intern this summer!

Publications

Powered by Jekyll and Minimal Light theme.