Marcio Cunha

Technical Articles

Practical articles, in-depth tutorials, and architectural insights on software engineering, AI, and systems.

All categoriesArquiteturaAutomaçãoBackendCarreiraDevOpsEletrônicaEngenhariaFrontendHardwareIANegóciosProdutividadeRedes
IABackend•4 min

Hallucination Mitigation in Language Models with Local Vector Databases

Learn how to safeguard your artificial intelligence applications against invented answers by leveraging local vector databases for strict fact verification.

IADevOps•6 min

Efficient Language Model Fine-Tuning with LoRA and 4-Bit Quantization on Local Clusters

Learn how to adapt large language models directly on your hardware using memory-saving techniques and distributed local processing.

IAArquitetura•4 min

Distributed Language Model Inference via Layer Splitting in Heterogeneous Clusters

Learn how to split massive language models across multiple different computers to run heavy artificial intelligence without needing expensive supercomputers.

IAArquitetura•4 min

Implementation of Context Retrieval in Knowledge Graphs for Error Minimization in LLMs

Learn how to structure knowledge graphs and retrieve contextual information to eliminate hallucinations in large language models by combining relational databases with generative AI.

IADevOps•5 min

Thermal Management in Edge Server Clusters with Reinforcement Learning

Learn how reinforcement learning algorithms optimize cooling in distributed edge server clusters, cutting energy costs and preventing hardware failures from overheating.

IAArquitetura•4 min

Hallucination Mitigation in Language Models with Sparse and Dense Vector Retrieval

Learn how to combine sparse and dense searches in Retrieval-Augmented architectures to eliminate language model hallucinations with surgical precision.

IAHardware•5 min

Language Model Inference Optimization on Low Power Hardware with 4 Bit Floating Point Quantization

Learn how to run massive language models on constrained, energy-efficient devices through 4-bit weight compression, maintaining precision without requiring powerful servers.

IA•4 min

Language Model Alignment with Reinforcement Learning and Human Preferences in Constrained Environments

Learn how to tune language models using reinforcement learning from human feedback on limited hardware, optimizing costs and operational efficiency.

IAArquitetura•4 min

Context Injection Attack Mitigation in Retrieval-Augmented Generation Systems

Learn how to protect your artificial intelligence application against malicious manipulations in retrieved data. Understand vulnerabilities in search and text generation architectures.

IABackend•4 min

Mitigating Language Model Hallucinations with Runtime Semantic Validation Layers

Learn how to implement semantic barriers and runtime validation to prevent artificial intelligence from generating falsehoods without losing conversational flexibility.