user@t8000:/$ cat publications.bib
Publications
Papers, conference talks and theses, newest first. Technical articles are listed below.
papers
- 2026 Zimbres, R.A. The Sliding Window Premium: Throughput, Goodput, and Per Layer Attention Costs of Hybrid Attention Models at Realistic Context Lengths on TPU v6e. Zenodo. [link]
- 2026 Zimbres, R.A. Batch Scaling and Goodput of a Tuned Attention Kernel on TPU v6e: Throughput, Latency, Energy, and Cost Measurements of vLLM Serving Gemma 4 31B. Zenodo. [link]
- 2026 Zimbres, R.A. The Decode Block Size Heuristic in TPU Ragged Paged Attention Reduces LLM Inference Throughput by 28 to 69 Percent. Zenodo. [link]
- 2026 Zimbres, R.A. From 1,540 to 19,511 Tokens per Second on a Single TPU Chip: A Measurement Study of LLM Inference Optimization on TPU v5e. Zenodo. [link]
- 2025 Zimbres, R.A.; Pappu, N. Privacy Challenges in Autonomous AI Agents. AI Agents Conference 2025, New York, USA.
- 2024 Zimbres, R.A.; Pappu, N. Mind the Data Gap: Privacy Challenges in Autonomous AI Agents. BlackHat Europe 2024, London, UK. [link]
- 2022 Zimbres, R.A. Cellular automata model of social networks (Version 1.0.0). CoMSES Computational Model Library. [link]
- 2010 Zimbres, R.A. Quality Perception Under a Dyadic Perspective: A Cellular Automaton Based Model. Academy of Management 2010, Montreal.
- 2009 Zimbres, R.A.; Oliveira, P.P.B. Dynamics of quality perception in a social network: A cellular automaton based model in aesthetics services. Electronic Notes in Theoretical Computer Science, Elsevier, v. 252. Also presented at the 15th International Workshop on Cellular Automata and Discrete Complex Systems. [link]
- 2009 Zimbres, R.A. The dynamics of quality perception in aesthetic services under a dyadic perspective. Doctorate thesis. [link]
- 2008 Zimbres, R.A.; Brito, E.P.Z.; Oliveira, P.P.B. Cellular automata based modeling of the formation and evolution of social networks: a case in Dentistry. ICEIS, International Conference on Enterprise Information Systems, Barcelona. [link]
- 2007 Zimbres, R.A. Game Theory and Transactions in Dentistry: A model based on cellular automata. RACE, Revista de Administração, Contabilidade e Economia, v. 4, p. 107-122. [link]
- 2007 Zimbres, R.A.; Brito, E.Z. Influence of rationality and quality in the purchase of services in an artificial society. SEMEAD, USP. [link]
- 2006 Zimbres, R.A. Effects of changes in the neighborhood and initial state in the flow of information in social networks. Wolfram Summer School. [link]
- 2006 Zimbres, R.A. Agent-Based Modeling: a Third Way of Doing Science? ENANPAD. [link]
- 2006 Zimbres, R.A.; Brito, E.Z. The use of Agent-Based Modeling in the interaction analysis of service clients. SIMPOI 2006, FGV-EAESP. [link]
- 2006 Zimbres, R.A. Exchanges of information in business networks. CATI, Annual Information Technology Congress (FGV). [link]
- 2005 Zimbres, R.A. The dynamics of the formation and evolution of business networks in Dentistry. Master's dissertation. [link]
- 2005 Zimbres, R.A. Negotiations in business networks based on Prospect Theory. Quantitative Methods and Informatics, SEMEAD, USP. [link]
evaluation board
- 2022 Gouvea, G.H. Natural language processing in the automatic analysis of resumes, videos and virtual interviews. Institute of Mathematics and Statistics, USP. [link]
articles
Tutorials and essays on Google Cloud, LLMs, agents, knowledge graphs and security. All of them: Medium · LinkedIn · Wolfram Community.
- Develop a Multi-Orchestration Neighborhood Watch App with Antigravity CLI
- Building a Healthcare Recommender with Keras, Two Towers and Google Cloud
- Developing a Variational Autoencoder in JAX using Antigravity
- Fine Tuning VaultGemma with Differential Privacy using a Colab Runtime in VSCode
- From Proof of Concept to Production: Building an Enterprise-Grade Platform for AI Systems
- Develop a Financial Multi-Agent System with Dynamic Tools using Gemini and Google ADK Agents
- Agent Development Kit: Enhancing Multi-Agent Systems with A2A Protocol and MCP Server
- Build an Xtreme Weather App with Google Geocoding and Places API
- Understanding Alzheimer's: Building Knowledge Graphs from Unstructured Data with Gemini
- Boost Your Website's Performance with Cloud Run Autoscaling
- Kolmogorov-Arnold Networks: a Critique
- In-Browser Inference with a Customized TensorFlow Lite Model and MediaPipe
- Multimodality with Gemini-1.5-Flash: Technical Details and Use Cases
- Use LLMs to Turn CSVs into Knowledge Graphs: A Case in Healthcare
- Fine Tuning Gemma-2b to Solve Math Problems
- Deploy a Generative AI Application with Terraform in Less Than 30 Minutes
- Cybersecurity in AI: Transfer Learning as an Attack Vector
- Building Knowledge Graphs from Scratch Using Neo4j and Vertex AI
- Deploying a Gemini Chatbot with Memory on a GKE Backend and Dialogflow
- Evaluating LLMs with LangChain: Using GPT-4 to Evaluate Google's Open Model Gemma-2B-it
- Cost-Efficient Multi-Agent Collaboration with LangGraph + Gemma for Code Generation
- Augmenting Gemini-1.0-Pro with Knowledge Graphs via LangChain
- SQL Queries + pgvector: Retrieval Augmented Generation in PostgreSQL
- How to Check if Your iPhone is Infected with Spyware/Stalkerware
- Code Generation using Retrieval Augmented Generation + LangChain
- Install a Real-Time Intrusion Detection System (IDS) with Suricata and Python
- LLM Cybersecurity Guide: OWASP Top 10 for LLM Apps
- Deploying a Google Cloud Generative AI App in a Website with Cloud Run
- Google Cloud for Education: Strategy Segmentation using Generative AI
- Setup of NVIDIA Merlin and TensorFlow for Recommendation Models
- Google Cloud Contact Center Artificial Intelligence (CCAI): A Managerial View
- Two Towers Recommender: A Custom Pipeline in Vertex AI Using Kubeflow
- Search of Brazilian Laws using Dialogflow CX and Matching Engine
- Graph Neural Networks: the Message Passing Algorithm
- Develop Secure End-to-End Machine Learning Solutions in Google Cloud
- Attacking Active Directory in a Windows Server Network with Kali Linux
- Agent-Based Modeling with Python, NetLogo and Arduino
- Burn a Physical Security Key to Access Your Google Accounts Securely
- Raspberry Pi IoT Project from the Command Line
- IoT on AWS: Dashboards from Streaming Sensor Data in Real Time
- GANs in TensorFlow from the Command Line
- Social Network Analysis Based on Calls Exchange
- Quantum States, CP Violation and Anti-Matter
- Markov Chain Monte Carlo to Save Computational Time