Anderson Gonçalves

Gameplay programmer and full stack engineer, MSc researcher at UFRGS

Half of my work has a budget measured per frame: gameplay systems in Unity and Unreal, some of them playable on this page. The other half has to stay up, and right now that means a compliance and logistics platform in Python and Angular with over a thousand automated tests. My MSc at UFRGS sits between the two. It is about control and decision-making, and NPCs are the test bench.

The three tracks

Featured game

All games
Controls: W thrusts, the mouse aims, left click fires.

Featured projects

All projects

Andy's Dungeon (opens in a new tab) Planned

A playable résumé, planned for Unity WebGL: professional experience and technical skills presented through a dungeon, its arcade machines and a grimoire CV.

UnityC#WebGLGame Design
Game Dev
★ 1

HTN + MORL Agent

MSc research — moving the learning to HTN method selection, so the agent adapts to changing priorities without rewriting the domain.

PythonReinforcement LearningHTN Planning
AI Research
Private repo

Game Insight (opens in a new tab)

Recommendation engine that unifies Steam, PlayStation and Xbox libraries and reads playtime and achievements, not genre tags. Led from ideation to working prototype under Program K (CEIA + EMBRAPII).

GoREST APIsRecommendation
Full Stack Game Dev
Go ★ 1

Latest publications

All publications

HTN Method Selection Guided by Multi-Objective Reinforcement Learning and Dynamic Preferences

Anderson G. A. C. Filho, Anderson R. Tavares, Lucas N. Alegre

In the proceedings of ERAMIA-RS 2026 II Escola Regional de Aprendizado de Máquina e Inteligência Artificial da Região Sul (SBC)

Four-page paper in the graduate track of ERAMIA-RS 2026, and the first version of this work. An HTN keeps the agent inside what the domain allows, and a preference-conditioned multi-objective RL policy picks among the valid methods according to the priorities of the moment. It has the formulation and the evaluation protocol. The experiments, and the full paper, come in 2027.

AI Research
9 Nov 2026 #decision-making#planning#reinforcement-learning

Teaching

Every unit I teach publishes its material here: slide decks, self-check exercises and the submission forms, open and in HTML. Units another teacher runs, or that I have not written yet, are listed too — so you can see the whole course, not only the part that is ready.

Open the material