Iván Sotelo

Data Scientist and AI enthusiast. Developing smart systems to achieve financial goals. #OpentoWork

[Iván Sotelo @ivansotelo12]

Great for any kind of agentic system or device application
@googlegemma

[Alibaba Cloud @alibaba_cloud]

Waiting on IT for data? That’s the bottleneck.
Quick BI Ad Hoc Analysis helps business users explore data faster — no SQL, no long waits. 🚀
Blog: https://t.co/iM9hzYxDOV 

Quick BI: https://t.co/1Kbo5iVhn1
#QuickBI #DataAnalytics #BI

[そう｜Claude Codeで始めるAI自動化 @so_ainsight]

セッションをまたいで記憶を保ち、経験から自分でスキルを作って育つAIエージェント Hermes Agent。ガチで使い倒すためのGitHubレポ10選：

21 Hermes Agent 本体
Nous Research公式のコアリポジトリ。MITライセンスで自由に使える。
https://t.co/Xj5kY3HUz4

2. Hermes-Wiki
Hermes Agentのソースコードを解説するコミュニティWiki。実装理解に効く。
https://t.co/6UGWzE6O3X

3. Atlas
エコシステム地図。100+のツール・スキルを俯瞰でき、RAG検索にも対応。
https://t.co/AuY36B2kkY

4. Control Interface
セルフホスト型ダッシュボード。複数エージェント・長時間タスク・記憶を一画面で管理。
https://t.co/XTHgrAfggT

5. Skill Factory
タスクを振り返って新スキルを自動生成・追加。エージェントが自分の武器を自製。
https://t.co/IvZ72W58kL

6. Maestro
ローカル動作のマルチエージェント協調ツール。Codex・Claude Code・Geminiを横断して構造化記憶と引き継ぎを管理。
https://t.co/X5h3q7uEPK

7. Hermes Agent Camel
信頼境界（CaMeL）を組み込んだフォーク版。本番運用の防護に向く。
https://t.co/vfSFVYi54T

8. Hermes HUD
TextualベースのTUI監視ターミナル。意識の流れとメモリ状態をリアルタイム可視化。
https://t.co/4IIEHRRNdY

9. Hermes Alpha
クラウド環境向けのHermes Agentデプロイ用テンプレート。Makefileと設定例同梱。
https://t.co/vgGr1aACLd

10. Awesome Hermes Agent
コミュニティ厳選のプラグイン・プロンプト・教材まとめリスト。
https://t.co/6ojreZi1yR

保存して順に試してみて。

[Nainsi Dwivedi @NainsiDwiv50980]

Learn AI for free directly from top companies.

1 - Anthropic:
https://t.co/33ebxYgBBp

2 - Google:
https://t.co/gI7pXjIu0V

3 - Meta:
https://t.co/IPnmhD24yq

4 - NVIDIA:
https://t.co/nnjmG2zAs0

5 - Microsoft:
https://t.co/E4rQyHTm7P

6 - OpenAI:
https://t.co/8WFoBtTzLz

7 - IBM:
https://t.co/7pYyKCI466

8 - AWS:
https://t.co/VvHGJHpBDt

9 - https://t.co/GUvkuEfOsS:
https://t.co/wDunExdaMF

10 - Hugging Face:
https://t.co/IugPdGNQOt

👇Comment "Learning" if you find this helpful.

Repost so others can take help.

Must bookmark for future reference.

[Paul Couvert @itsPaulAi]

Xiaomi has released an open source model "MiMo-V2.5-Pro"… and it’s SO GOOD for agents and coding 🔥

You can plug it into OpenClaw, Hermes Agent, Claude Code, and more.

Might be one of the best options for frontend tasks as well

Example here with a one-shot 3D game optimized for mobile (more below)

[GitHub Projects Community @GithubProjects]

Awesome Agent Skills: The Largest Curated Library of AI Agent Skills.

[Visual Studio Code @code]

Install Agent Plugins in VS Code to add skills, commands, MCP servers, hooks, and more ⚡

[千寻｜AI 分享 🌸 @Crypto_QianXun]

如果我要在90天内找到一份20万美元的AI工程师工作，我不会去读学位。

我会精通这10个GitHub仓库。

1. awesome-llm-apps
生产级AI指南。RAG、智能体、多模态应用，附完整代码。10.6万+ stars。
仓库 → https://t.co/0ZZKDLflhd

2. LangChain
基础框架。被Klarna、Replit、Elastic及2026年大多数AI创业公司用于生产环境。
仓库 → https://t.co/XqfpGJReBi

3. LangGraph
驱动生产智能体的编排层。资深AI工程师岗位描述中的必备技能。
仓库 → https://t.co/aHDtWGRsF0

4. CrewAI
多智能体协调。财富500强团队首选框架。
仓库 → https://t.co/JFbdjgk4Mt

5. Ollama
在自己机器上运行任何开源LLM。学习模型工作原理的最快方式。
仓库 → https://t.co/yySYBNIw7X

6. awesome-mcp-servers
MCP是2026年所有主流AI实验室采用的标准。掌握它让你领先99%的工程师。
仓库 → https://t.co/O6zfXs66cK

7. Qdrant
大规模生产RAG使用的向量数据库。嵌入和语义搜索是AI岗位的必需技能。
仓库 → https://t.co/mRXnbqF2XP

8. AI-Agents-for-Beginners
微软免费12课程，教建立智能体。真实代码、真实练习、真实面试准备。
仓库 → https://t.co/nbhJglVFHl

9. system-design-primer
生产级AI就是系统设计。FAANG工程师用来面试准备的仓库。
仓库 → https://t.co/ojgvZFVG25

10. awesome-claude-code
这个工具的使用指南，现被FAANG、OpenAI、Anthropic及大多数YC创业公司内部使用。
仓库 → https://t.co/D8EjUYiz3w

[Kanika @KanikaBK]

🚨 THIS GITHUB REPO JUST BECAME MY MOST USED BOOKMARK OF 2026.

1000+ free AI tools. One place. Updated every week.

One GitHub repo. Every open source AI tool that actually matters.

Organized. Updated. Free.

Here is what is inside:
↳ The exact models companies like Google, Meta, Microsoft, and Alibaba use internally, but open source and free
↳ Tools to run AI on your own laptop with zero subscription
↳ Frameworks to build your own ChatGPT, your own image generator, your own AI agent
↳ The same inference engines that power production apps serving millions of users
↳ 200+ AI models for text, image, video, audio, and 3D, all free
This is not a beginner list. This is what engineers at actual AI companies use.

And it is all sitting there. Free. Right now.

The categories alone will make your head spin:
↳ AI models you can run locally with no internet
↳ Tools to build AI agents that actually do things
↳ Open source alternatives to Midjourney, Sora, ElevenLabs, and Cursor
↳ Full courses to go from zero to building real AI products
↳ Security and safety tools the big labs use internally

If you are paying for any AI tool right now, there is a 70% chance there is a free open source version in this repo that does the same thing.

The people who find this list in 2026 and actually use it are going to build things that look impossible to everyone else.

Save this post. Open the repo. Pick one tool. Start today.

LINK IN COMMENT

[そう｜Claude Codeで始めるAI自動化 @so_ainsight]

90日でAIエンジニアになりたいなら、オンライン講座より先にやることがある。
このGitHubリポジトリ10個から1つ選んで、手を動かすこと。

1. LangChain
ほぼ全AIエンジニア求人で名前が出るLLMアプリのフレームワーク。本番LLMアプリを作るならここから。
https://t.co/GOOh9JB56P

2. LangGraph
ステートフルなエージェントをグラフで設計。求人の「エージェントワークフロー」はだいたいこれ。
https://t.co/eLWEIkFvQt

3. LlamaIndex
RAGとドキュメントエージェントの定番。求人の「検索パイプライン」はここを指してる。
https://t.co/h0LqgrvUXN

4. CrewAI
役割とタスクで動くマルチエージェントチーム。Fortune 500企業の本番環境で稼働中。
https://t.co/AklNDTXjUf

5. Qdrant
Rust製の本番向けベクトルDB。求人ではPinecone・Chroma・FAISSと並んで名前が挙がる。
https://t.co/IONnvtDwgz

6. Ragas
RAGパイプライン評価の標準フレームワーク。ハルシネーション・忠実性・関連性が数値で測れる。
https://t.co/eQs8OmWWxz

7. Ollama
オープンソースLLMをコマンド1発でローカル実行。コストとプライバシー面で求人需要が増えてる。
https://t.co/wvdidJsTO9

8. Awesome MCP Servers
MCP（Model Context Protocol）は求人で一番新しいスキル。本番運用中のMCPサーバーが網羅されてる。
https://t.co/1Yf8Kh9AC2

9. Awesome LLM Apps
RAG・エージェント・マルチエージェント・音声・MCP用のテンプレが100以上。動くコードがそのまま手に入る。
https://t.co/RqNrlzE7wD

10. AI Agents for Beginners
Microsoft公式の無料12レッスン。AIエージェントの全体像を有料壁・登録なしで学べる。
https://t.co/OSseMAybLN

2026年のAIエンジニア求人で求められるのは、RAG・エージェント・ベクトルDB・評価・MCP。
この10個で全部触れる。

1個選んで、1プロジェクト作って、GitHubに上げる。スタートはそれで十分。

全部無料、全部オープンソース。

[Tableau @tableau]

Your AI is only as good as the data it can access.

Transform any agent into a data expert with Tableau MCP. Securely query Tableau’s analytics engine directly for accurate, grounded insights: https://t.co/Y5sRnnXEts

[DailyPapers @HuggingPapers]

MSA breaks the 100M token barrier

Memory Sparse Attention achieves unprecedented 100M token context lengths with near-linear complexity. The architecture maintains 94% accuracy at 1M tokens while outperforming RAG systems and frontier models, using end-to-end sparse attention with document-wise RoPE.

[homanp @pelaseyed]

Woow, someone made a desktop app for infinite monitor.

[0xMarioNawfal @RoundtableSpace]

MIROSHARK LETS YOU SIMULATE HOW THE INTERNET REACTS TO ANY DOCUMENT - BEFORE YOU PUBLISH IT.

Upload a press release, policy draft, or financial report and it generates hundreds of AI agents with unique personalities grounded in a knowledge graph.

Here's what it actually does:

> Extracts entities and relationships from your document into a knowledge graph
> Generates agent personas with 5 layers of context each
> Simulates reactions simultaneously across Twitter, Reddit, and Polymarket
> Agents see cross-platform context - traders read social posts, social agents see market prices
> A ReACT agent writes analytical reports on what agents said and how markets moved
> Chat directly with any agent or send questions to groups

The simulation tracks belief states, confidence, and trust per agent updating every round.

PR teams, policy makers, and analysts spend thousands stress-testing public reaction.

This runs locally, with any cloud API, for free.

GitHub: https://t.co/WXzJAvRI6q

[Tom Dörr @tom_doerr]

Self-hosted AI runtime for OpenClaw

https://t.co/hHzcS9VWcP

[Hugging Models @HuggingModels]

Meet Gemma-2-2B-IT: a lightweight conversational AI that's making waves. With 2 billion parameters, it's surprisingly capable for its size. Perfect for devs who want quality text generation without massive compute requirements. Why's everyone excited? It punches way above its weight class.

[jessica🩶 @ds_jessica_]

Generative AI for Software Development Skill Certificate

🚀 Introduction to Generative AI for Software Development
🤝 Team Software Engineering with AI
🤖 AI-Powered Software and System Design
https://t.co/FB8LmAbSUI

[Tom Dörr @tom_doerr]

AI generated architecture docs for humans

https://t.co/hG9jii8q6w

[Nav Toor @heynavtoor]

🚨 An AI just wrote a scientific paper.

Came up with the hypothesis. Designed the experiments. Ran the code. Analyzed the data. Created the figures. Wrote every word.

Then it passed peer review at a top machine learning conference.

No human touched it. Not one word. Not one edit.

This is not a demo. This actually happened. At ICLR 2025.

It's called AI Scientist v2.

An open source system that does the entire scientific research process. Autonomously. End to end. From idea to published paper.

Here's what this system does on its own:

→ Generates research hypotheses from a broad topic you provide
→ Searches existing literature to check if the idea is novel
→ Designs experiments to test the hypothesis
→ Writes and debugs its own experiment code
→ Runs the experiments on GPUs
→ Analyzes the results with statistical methods
→ Creates publication-ready figures and visualizations
→ Writes the entire manuscript. Title to references. LaTeX formatted.
→ Reviews its own paper and improves it before submission

Here's the wildest part:

They submitted 3 fully AI-generated papers to an ICLR workshop. Reviewers were told some papers might be AI-generated but not which ones. One paper scored 6, 7, and 6 from three reviewers. That put it in the top 45% of all submissions. Above the average human paper.

The AI outscored most human researchers. At a real conference. Through blind peer review.

PhD programs cost $50,000 to $80,000 per year. Research takes 5 to 7 years. Postdocs earn $55,000 for more years of the same grind.

2.2K GitHub stars. Published research paper. Apache 2.0 License.

100% Open Source.

[GitHub @github]

🆕 The Awesome GitHub Copilot project has a new home.

Head over to explore hundreds of community-built customizations:
🔍 Full-text search for agents and skills
📚 A dedicated Learning Hub
⚡ 1-click plugin installs for Copilot CLI & @code

Built by the community, for the community. Check it out.👇
https://t.co/OUL39XaFq4