Keynote Speakers

Roberto Ierusalimschy

Honorary Visiting Professor, PUC-Rio

Bio: Roberto Ierusalimschy is the chief architect of the Lua programming language. Until July 2025, he was a Full Professor at PUC-Rio, where he is now an Honorary Visiting Professor.

Small is Beautiful: the Design of Lua

Lua is a programming language developed at the Catholic University in Rio de Janeiro that came to be the leading scripting language in video games. Lua is also used extensively in embedded devices, such asset-top boxes and TVs, and other applications like Adobe Lightroom and Wikipedia.

Along the years, the design of the Lua has kept simplicity and smallness among its main goals, within the restraints of a language with a large installed user base. In this talk we will discuss the design of Lua, emphasizing the choices and tradeoffs involved in keeping the languagesimple and small while making it a unique fit for its many applications.

Emily Short

Game Designer, DeepMind

Bio:

Emily Short works in interactive fiction, narrative design, and conversation modeling. She is a game designer at DeepMind and was previously narrative director at Wizards of the Coast, and creative director for three years at Failbetter Games, a British studio focused on narrative. She has released over two dozen solo games, as well as contributing to indie games such as Sunless Sea, Sunless Skies, Orwell's Animal Farm, Zombies, Run, and Where the Water Tastes Like Wine, and also frequently consults with studios of all sizes.

Emily Short's AI experience includes three years as a product manager and then CPO at Spirit AI, where she worked with natural language processing and generation, character moods and social behavior, and conversation modelling. She also served for three years as an advisor to the GDC AI Summit, and was previously the creative director of the Versu project, building interactive iPad stories around AI character agents.

Her blog can be found at http://emshort.blog.

Testing-First Tools for Developing Procedural Narratives

This talk concerns personal research work, and does not reflect the opinions or research of any of Emily Short's current or previous employers.

Procedural narrative games often rely on a collection of human-written story elements with data-driven conditions for how they can be selected and recombined. Such systems tend to be highly flexible and extensible, as well as easy to combine with non-narrative gameplay, but their behavior depends on properties of the data set as a whole, which can be challenging to plan, reason about, and tune.

We'll look at a project that used LLM coding tools (primarily Claude) to rapidly prototype narrative tooling. This tooling allows a designer to define the qualities of the intended system; generate a body of stub data to meet its requirements; and simulate the system behavior to test whether the intent has been met, using player policies to simulate different gameplay approaches. The result is a game runtime free of LLM involvement and relying entirely on human authorship, but which does not impose as large a development burden on the human designer as more typical techniques.

The project also surfaces some generalised design vocabulary for describing the features of storylet and similar procedural narrative systems.

Gonzalo Frasca

Game Designer, Independent Scholar

Bio:

Gonzalo Frasca is a writer and game designer who works in Game Learning R&D at Kahoot!. A pioneering scholar in the field of game studies, he has designed games for leading entertainment studios, including Cartoon Network, Pixar, Disney, Warner Bros., and Lucasfilm.

Frasca holds a PhD from the Center for Computer Games Research at the IT University of Copenhagen (ITU), with research focused on game rhetoric and political game design. He is the former Chair of Video Games at Universidad ORT Uruguay, the first university-level game design program in Spanish-speaking Latin America.

Recently, he co-created DragonBox School, a full-curriculum, game-based math learning solution used across Finland, France, and Norway.

Frasca received a Distinguished Alumni Pioneer Award from the Georgia Institute of Technology and a Lifetime Achievement Award from the Knight Foundation for his pioneering work in newsgames and game-based journalism.

Borges, Games, and Artificial Intelligence

Unlike The Simpsons, Argentine writer Jorge Luis Borges was not in the business of predicting the future. Yet his playful stories, poems, and conceptual games anticipated AI’s most fascinating questions: memory, perception, intelligence, authorship, and how language constructs reality.

Let’s follow Borges into the labyrinth of his writings to see what we can learn about the challenges—and creative fun—of Artificial Intelligence.

Page updated

Google Sites

Report abuse