Matthew

1.3K posts

Matthew

Matthew

@Matthew155261

Katılım Mart 2025
201 Takip Edilen25 Takipçiler
Matthew retweetledi
العربية
العربية@AlArabiya·
كردستان العراق.. قوات التحالف أسقطت مسيرات في سماء أربيل
العربية tweet media
العربية
0
6
27
12.5K
Matthew retweetledi
Cagatay Ulusoy
Cagatay Ulusoy@ulusoyapps·
6 months ago I asked this question to the @FlutterDev team on #FlutterFlightPlans YouTube live stream: "Are there any use cases for generative UI other than chat responses?" and the answer from @sethladd was "Let's see what is beyond chat together" 🚀 Since then, I have been exploring interesting use cases with GenUI. Happy to share that I am now starting another very interesting GenUI article series 👀 stay tuned!
Cagatay Ulusoy tweet media
English
1
3
22
999
Matthew retweetledi
Big Technology Podcast
Big Technology Podcast@BigTechPod·
Jürgen Schmidhuber (@SchmidhuberAI) on the AI infrastructure trap: "If you invest $1 trillion today into GPUs for data centers, within five years you're going to lose $900 billion. Somebody is going to lose $900 billion in the near future. Because there is no business model to recuperate all these losses."
English
13
26
135
41.7K
Matthew retweetledi
WOLF Bitcoin
WOLF Bitcoin@WOLF_Bitcoin_·
GOLD IS DOWN ALMOST -2% 🔴🔴 AND DROPS BELOW $4,000/OZ
WOLF Bitcoin tweet media
English
4
2
10
2K
Matthew retweetledi
Dan Kornas
Dan Kornas@DanKornas·
AI coding is less useful when you spend more time moving context around than reviewing changes. Aider is an AI pair-programming tool in your terminal for developers starting a project or working in an existing codebase. It helps you make repository-aware code changes by mapping your codebase, editing files, and integrating each change with Git. Key features: • Codebase map – gives the model a map of your repository to support work across larger projects • Cloud and local models – connects to almost any LLM, including local options • Git-native workflow – automatically commits edits so you can diff, review, manage, or undo changes • Lint and test loop – runs configured linters and tests after edits and can fix reported problems • Extra context – accepts images and web pages alongside your code and instructions It’s open-source (Apache 2.0 license). Link in the reply 👇
Dan Kornas tweet media
English
4
4
14
1.4K
Matthew retweetledi
العربية
العربية@AlArabiya·
ألمانيا نقترح إرسال بعثة أوروبية محل قوات "اليونيفيل" إلى لبنان
العربية tweet media
العربية
1
3
7
11.2K
Matthew retweetledi
Dan Kornas
Dan Kornas@DanKornas·
Connecting an agent to real apps means handling tool discovery, authentication, and per-user state—not just function calls. Composio is an SDK monorepo for builders who need AI agents to take actions across external apps. It helps you add app tools by creating user-scoped sessions that discover, authenticate, and execute tools at runtime without loading hundreds of tool definitions into context. Key features: • 1000+ app toolkits – give agents access to actions across supported apps • Per-user sessions – scope tools and connected accounts to individual users • Runtime meta tools – discover, authenticate, and execute tools when needed • TypeScript and Python SDKs – build with either supported language stack • Framework and MCP support – use provider adapters or a hosted session endpoint The SDK repository is open-source (MIT license); the quickstart requires a Composio API key. Link in the reply 👇
Dan Kornas tweet media
English
1
1
4
1.3K
Matthew retweetledi
Python Programming
Python Programming@PythonPr·
SQL Essential Commands Every Developer Should Know!
Python Programming tweet media
English
0
33
167
6.3K
Matthew retweetledi
Kirk Borne
Kirk Borne@KirkDBorne·
"Machine Learning Engineering on AWS: Build, deploy, and operationalize LLMs, AI agents, and generative AI systems on AWS" — amzn.to/4od7C1I 𝗪𝗵𝗮𝘁 𝗬𝗼𝘂 𝗪𝗶𝗹𝗹 𝗟𝗲𝗮𝗿𝗻: 🟢Build and deploy AI agents using Bedrock AgentCore and Strands Agents 🟡Dive deep into ML engineering with Amazon SageMaker AI 🟠Evaluate model performance using LLM-as-a-judge 🟢Explore advanced model fine-tuning and deployment using SageMaker AI 🟡Build RAG-powered systems using Bedrock Knowledge Bases and S3 Vectors 🟠Modernize analytics with a managed transactional data lake 🟢Automate LLMOps pipelines using SageMaker Pipelines and AWS Lambda 🟡Explore best practices for building GenAI systems and AI agents on AWS 🟠SageMaker AI Model Training, Tuning Capabilities, Deployment Options, and Strategies This book is intended for AI engineers, data scientists, machine learning engineers, and technology leaders who want to deepen their understanding of machine learning engineering, generative AI, large language models, retrieval-augmented generation, AI agents, and MLOps on AWS.
Kirk Borne tweet media
English
1
4
17
2.2K
Matthew retweetledi
David Gokhshtein
David Gokhshtein@davidgokhshtein·
win together
English
59
25
433
47.4K
Matthew
Matthew@Matthew155261·
@AlArabiya What's the big news about Trump declassifying somethingd.
English
0
0
0
19
العربية
العربية@AlArabiya·
ترامب: سنرفع السرية عن وثائق بشأن الانتخابات الأميركية ara.tv/w96ya
العربية
2
4
13
13.5K
Matthew
Matthew@Matthew155261·
@Ronald_vanLoon @verge Interesting approach to minimalism with wearable tech! Less is indeed more in today's fast-paced world.
English
0
0
1
5
Matthew
Matthew@Matthew155261·
AI algorithms accurately identify suspicious transactions, ensuring that no one can illegally claim subsidies and escape detection.
Matthew tweet media
English
0
0
0
1
Matthew
Matthew@Matthew155261·
@DanKornas "Just another day in the world of AI and ML, huhd".
English
1
0
1
23
Dan Kornas
Dan Kornas@DanKornas·
Your docs should answer questions, not make users hunt through pages. AutoFlow is an open-source GraphRAG knowledge base tool for teams building conversational search over product and documentation content. It helps you turn documentation into a searchable knowledge base by crawling sitemap URLs and indexing content with vector and optional knowledge-graph indexes. Key features: • Conversational search – provides a built-in, Perplexity-style interface for asking questions. • Website ingestion – imports pages through sitemap.xml or individual URLs. • File ingestion – accepts Markdown, PDF, Word, PowerPoint, Excel, and text files. • GraphRAG indexing – supports vector indexes plus optional knowledge-graph indexes with entities and relationships. • Embeddable widget – adds the conversational search window to a site with a JavaScript snippet. It’s open-source (Apache 2.0 license). The README also says AutoFlow is still in the early stages of development. Link in the reply 👇
Dan Kornas tweet media
English
3
3
26
1.7K