Sergio De Simone

Barcelona

Editor at InfoQ

Featured in: Favicon

infoq.com

dzone.com

newsbreak.com Favicon

thehoopsnews.com Favicon

codeproject.com

Articles

Google Releases LMEval, an Open-Source Cross-Provider LLM Evaluation Tool

1 week ago | infoq.com | Sergio De Simone

LMEval aims to help AI researchers and developers compare the performance of different large language models. Designed to be accurate, multimodal, and easy to use, LMEval has already been used to evaluate major models in terms of safety and security. One of the reasons behind LMEval is the fast pace at which new models are being introduced. This makes it essential to evaluate them quickly and reliably to assess their suitability for specific applications, says Google researchers.
Gemma 3n Available for On-Device Inference Alongside RAG and Function Calling Libraries

1 week ago | infoq.com | Sergio De Simone

Google has announced that Gemma 3n is now available in preview on the new LiteRT Hugging Face community, alongside many previously released models. Gemma 3n is a multimodal small language model that supports text, image, video, and audio inputs.
Meta Open-Sources Pyrefly, a High-Performance Python Type Checker in Rust

1 week ago | infoq.com | Sergio De Simone

Currently in alpha, Pyrefly is a new open-source Python type checker developed by Meta in Rust for maximum performance. Inspired by tools like Pyre, Pyright, and MyPy, Pyrefly is intended to replace the OCaml-based Pyre type checker previously used for Instagram’s codebase. We built a custom engine for incremental computation and designed our type-checking algorithm based on years of experience in gradual typing theory and Rust expertise.
Google Enhances LiteRT for Faster On-Device Inference

2 weeks ago | infoq.com | Sergio De Simone

The new release of LiteRT, formerly known as TensorFlow Lite, introduces a new API to simplify on-device ML inference, enhanced GPU acceleration, support for Qualcomm NPU (Neural Processing Unit) accelerators, and advanced inference features.
Docker Introduces Hardened Images to Strengthen Container Security

2 weeks ago | infoq.com | Sergio De Simone

Docker has launched Docker Hardened Images, a catalog of enterprise-grade, security-hardened container images designed to protect against software supply chain threats. By relieving DevOps teams from the chore of securing their containers on their own, hardened images provide an easier way to meet enterprise-grade security and compliance standards, Docker says. Hardened Images aim to increase teams' confidence that image components have not been tampered with and do not include malicious code.
Llama 4 Scout and Maverick Now Available on Amazon Bedrock and SageMaker JumpStart

3 weeks ago | infoq.com | Sergio De Simone

AWS recently announced the availability of Meta's latest foundation models, Llama 4 Scout and Llama 4 Maverick, in Amazon Bedrock and AWS SageMaker JumpStart. Both models provide multimodal capabilities and follow the mixture-of-experts architecture. Launched by Meta last April, Llama 4 Scout and Maverick include 17 billion active parameters distributed across 16 and 128 experts, respectively. Llama 4 Scout is optimized to run on a single NVIDIA H100 GPU for general-purpose tasks.
Shopify社とReact Nativeの歩み：5年を駆けるレトロスペクティブ

3 weeks ago | infoq.com | Sergio De Simone

InfoQ ホームページニュース Shopify社とReact Nativeの歩み：5年を駆けるレトロスペクティブ原文リンク(2025-04-17) 5年にわたるReact Native開発を振り返り、Shopify社のモバイルチームが開発エクスペリエンスから得られた貴重な洞察した。開発過程における成功事例、失敗事例、今後の展望に焦点を当てた内容である。 Spotify社のMustafa Ali氏は、自社エンジニアのReact Nativeの実用エクスペリエンスから、いくつかの知見が得られたと述べている。ユーザーにとってもっともありがたい点は、React Nativeで構築されたアプリケーションは厳格な実行処理要求にも応答が可能な点だろう。だが、成功を左右するのは適切なパターンの効果的な適用であり、タネや仕掛けは存在しないのだと同氏は指摘している。わが社では自社アプリのロード画面時間を500ms以下（P75）に削減しました。ネイティブ開発と同様に、React Native開発でも適切なパターンと技術を用いたボトルネックの解消が要求されます。...
Meta Open Sources LlamaFirewall for AI Agent Combined Protection

3 weeks ago | infoq.com | Sergio De Simone

LlamaFirewall is a security framework aimed at safeguarding AI agents against prompt injection, goal misalignment, and insecure code generation. It achieved over 90% efficacy in reducing attack success rates when evaluated on the AgentDojo benchmark. Additionally, developers can update its behavior by adding new security guardrails. LlamaFirewall is a real-time guardrail monitor designed to serve as a final layer of defense against security risks associated with AI Agents.
Android Studio Meerkat Update Enhances Gemini Integration and More

4 weeks ago | infoq.com | Sergio De Simone

The latest feature drop for Android Studio Meerkat brings a range of updates aimed at boosting developer productivity, including extended Gemini integration for crash analysis and unit testing, enhanced Jetpack Compose previews, better Kotlin Multiplatform (KMP) integration, and more. Gemini integration in Meerkat now helps developer to investigate, understand, and resolve crashes by generating insights into their root causes.
Meta Launches AutoPatchBench to Evaluate LLM Agents on Security Fixes

1 month ago | infoq.com | Sergio De Simone

AutoPatchBench is a standardized benchmark designed to help researchers and developers evaluate and compare how effectively LLM agents can automatically patch security vulnerabilities in C/C++ native code. AutoPatchBench comprises a collection of tests aimed at evaluating the ability of LLMs to autonomously generate security patches for vulnerabilities identified using fuzz testing.