Fernando Ike
fernandoike.bsky.social
Fernando Ike
@fernandoike.bsky.social
"Why open source may not survive the rise of generative AI | ZDNET" share.google/hX1MWxsQgHod...
Why open source may not survive the rise of generative AI
Generative AI may be eroding the foundation of open source software. Provenance, licensing, and reciprocity are breaking down.
share.google
October 25, 2025 at 11:29 PM
Ótimo texto do Russ Cox sobre Supply-Chain, Segurança, OpenSource e backdoors. E um pouco de história sobre backdooreem 1974...

Leitura para o cafézinho...

dl.acm.org/doi/10.1145/...
Fifty Years of Open Source Software Supply-Chain Security | Communications of the ACM
dl.acm.org
October 18, 2025 at 12:58 PM
Reposted by Fernando Ike
October 7, 2025 at 2:24 AM
"Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory"

arxiv.org/abs/2509.14662
Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory
While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a nov...
arxiv.org
September 29, 2025 at 9:31 AM
Reposted by Fernando Ike
Amanhã devo soltar um blog post sobre o kyaml, pq ele é uma feature foda do Kubernetes e como essa preocupação que tenho visto com o formato ser verboso demais não condiz com a realidade :P me aguardem, cheio de ironia e amor

E podia ser pior, podia ser xml então parem de reclamar :x
August 29, 2025 at 1:17 AM
"A new perspective on how cosmological correlations change based on kinematic parameters"

share.google/l4lSzUdEAgmb...
A new perspective on how cosmological correlations change based on kinematic parameters
To study the origin and evolution of the universe, physicists rely on theories that describe the statistical relationships between different events or fields in spacetime, broadly referred to as cosmo...
share.google
August 24, 2025 at 8:56 PM
Gosta de um bom post-mortem com conceitos, fundamentos e profundamente detalhado?

São pouco mais de 100 páginas, mapas, ilustrações, imagens para explicar as causas de uma colisão de um avião e um veículo na pista do Aeroporto do Galeão em fevereiro de 2025.
sistema.cenipa.fab.mil.br
August 10, 2025 at 12:23 AM
Vale uma lida no Al Action Plan da Casa Branca, há uns ponto de destaque ao considerar ações de longo prazo para uma nação ou mesmo empresas...

Americas-Al-Action-Plan.pdf Inkd.in/dZKqjDDe
August 2, 2025 at 6:43 PM
"A Knockout Blow for LLMs?
Anybody who thinks LLMs are a direct route to the sort of AGI that could fundamentally transform society for the good is kidding themselves"

cacm.acm.org/category/art...
Artificial Intelligence and Machine Learning – Communications of the ACM
cacm.acm.org
June 17, 2025 at 9:14 AM
Critical Langflow Vulnerability (CVE-2025-3248) Actively Exploited to Deliver Flodrix Botnet | Trend Micro (US) share.google/lpotLhk7pAVq...
Critical Langflow Vulnerability (CVE-2025-3248) Actively Exploited to Deliver Flodrix Botnet
This blog uncovers an active campaign exploiting CVE-2025-3248 in Langflow versions before 1.3.0 that deploys the Flodrix botnet, enabling threat actors to achieve full system compromise, initiate...
share.google
June 17, 2025 at 7:21 AM
Infrastructure From Code: What Went Wrong - The New Stack share.google/HqK0JJ3UEn9O...
Infrastructure From Code: What Went Wrong
Developers were reluctant to give up control and vendors struggled to meet diverse compliance and operational needs. Now, the focus is shifting.
share.google
June 16, 2025 at 10:30 AM
[2506.08837] Design Patterns for Securing LLM Agents against Prompt Injections share.google/s0Eh2WQsP29H...
Design Patterns for Securing LLM Agents against Prompt Injections
As AI agents powered by Large Language Models (LLMs) become increasingly versatile and capable of addressing a broad spectrum of tasks, ensuring their security has become a critical challenge. Among t...
share.google
June 14, 2025 at 3:59 PM
Reposted by Fernando Ike
Bridge Types

xkcd.com/3097/
June 3, 2025 at 6:45 PM
WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference

arxiv.org/abs/2505.19427
WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference
The growing computational demands of large language models (LLMs) make efficient inference and activation strategies increasingly critical. While recent approaches, such as Mixture-of-Experts (MoE), l...
arxiv.org
June 3, 2025 at 10:48 AM
"TinyV: Reducing False Negatives in Verification
Improves RL for LLM Reasoning"

arxiv.org/pdf/2505.14625
arxiv.org
May 25, 2025 at 1:05 AM
Saudosismo de fine tuning do velhos msdos...

github.com/microsoft/edit
May 21, 2025 at 5:50 PM
"LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities"

arxiv.org/abs/2504.160...
LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities
The success of Large Language Models (LLMs) has sparked interest in various agentic applications. A key hypothesis is that LLMs, leveraging common sense and Chain-of-Thought (CoT) reasoning, can effec...
arxiv.org
May 21, 2025 at 3:06 AM
Texto da Codecon sobre DORA e SPACE que vale vc ler. Se gostar do texto é porque o Gabriel Nunes é um ótimo perguntador, se não gostar é porque algumas respostas são minhas. 😅

"Métricas de Performance DORA e SPACE: medindo a produtividade de devs com exemplos reais"

codecon.dev/blog/m%C3%A9...
Métricas de Performance DORA e SPACE: medindo a produtividade de devs com exemplos reais - Blog - Codecon
Entenda como aplicar as métricas DORA e SPACE no dia a dia da profissão.
codecon.dev
May 11, 2025 at 11:39 PM