Reindent (@ReindentAI) header
Reindent (@ReindentAI) profile picture

We build AI agent teams that run work on their own. Reindent runs on them: software development, research, videos, posts and security, all on https://nitter.cf/t.co/ojhWnbFh4B

United States
Joined January 2024
Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)Photo by Reindent (@ReindentAI)
Reindent retweeted
Hackers have AI agents now. So do I. Two workflow strategies my agents run on their own to protect my web apps: 1) Proactive: scout X several times a day for new vulnerabilities, rate the risk 0-10, alert me. 2) Reactive: watch my servers for anomalies. If there's a breach, take that service offline (no data deleted) and email me until I answer. I have some more I'll share soon. All running on @jauvex
1
1
2
101
Reindent retweeted
Today: 70% of the web runs on PHP Next year: 70% of the web runs on vibe coded apps AI cybersecurity will be such an important field in the coming years, I already incorporated cybersecurity agents in my AI workflows, this is a must for anyone running agentic companies.
1
1
1
134
Reindent retweeted
This is a new personal record: my video agents managed to autonomously render a video about @Tesla's new flying Roadster, from edit to upload in 30 mins. All thanks to @jauvex! We're truly living in the future... flying cars!!
1
1
2
104
Reindent retweeted
Jauvex 1.3 released! Now including automated workflows. Import your current Claude, Codex or Grok sessions, and ask your agents to turn their current tasks into consistent, automated workflows or routines that Jauvex executes for you! Fully open-source. Links in thread.
1
1
65
Reindent retweeted
This is something that should be native in @claudeai @ChatGPT @grok and other agents A way for agents to talk to each other, hand off tasks, or reach consensus. /agent-talk Codex, Claude says you're smelly 4 minutes later: CONSENSUS REACHED. Codex agrees he is smelly. agent-talk.reindent.com Just released, read more 🧵👇
2
2
2
82
Reindent retweeted
AI agents are vulnerable to a new kind of attack: prompt injection. Models see everything in prompts. If an attacker controls any part of them, they can alter the behavior of your agent. Assume compromise. Limit tool call access. Don't trust the output. vercel.com/blog/building-sec…
15
63
8
444
63,233
Reindent retweeted
GPT-4.1 (and -mini and -nano) are now available in the API! these models are great at coding, instruction following, and long context (1 million tokens). benchmarks are strong, but we focused on real-world utility, and developers seem very happy. GPT-4.1 family is API-only.
730
876
320
11,058
1,621,872
Below, we have an MCP-driven Agentic RAG that searches a vector database and falls back to web search if needed. To build this, we'll use: - Bright Data to scrape web at scale. - @qdrant_engine as the vector DB. - @cursor_ai as the MCP client. Let's build it!
5
8
70
14,483
Scrapybara is transitioning from providing virtual desktop environments for AI agents to becoming a dedicated AGI research lab. Our mission is simple: develop AGI systems solving humanity's biggest challenges: world hunger, cancer, and unemployed CS grads. Join us.
18
4
3
165
13,669
Reindent retweeted
How would @deepseek_ai models x @OpenAI operator look like? Wait no more, @nanobrowser_ai now supports @deepseek_ai official API, including the most recent released DeepSeek V3-0324! Get the new release, use it right now in your own browser 👇
5
3
2
22
11,276
Reindent retweeted
4o image generation has arrived. It's beginning to roll out today in ChatGPT and Sora to all Plus, Pro, Team, and Free users.
952
1,826
829
20,175
4,924,615
This not only has aged well, it’s also the single most important change that’s happened to the software engineering industry since its inception
The hottest new programming language is English
73
236
14
3,872
401,747
Agentic commerce is...actually...starting... You can now buy [coffee beans] via an AI agent using @stripe. 1/ Find [coffee] with AI 2/ Buy in 1-click with Link 3/ Agent gets one-time virtual card 4/ Agent purchases via @browserbase ☕ Watch this video (built by @EvanFenster)
54
54
16
565
130,131
🔍 Build a Deep Research Agent An autonomous research agent built with LangGraph that combines GPT-4 and Tavily AI to perform parallel research and generate structured reports, all through sophisticated state management. 🚀 Check out the implementation: analyticsvidhya.com/blog/202…
6
170
7
685
46,661
Reindent retweeted
I am getting ready to teach "Lesson 1: Agent" to 3000 students at 8am PST tomorrow. Join my 4-week mini-course: byhand.ai/p/introduction-to-… Lesson plan 👇
3
16
3
152
14,628
The future of AI is here and it's not from Silicon Valley. Manus AI just launched from China and it's blowing every other AI agent out of the water. What makes Manus special: • Browser control that works flawlessly • File creation and editing directly in chat • Website deployment from simple prompts • Lightning-fast research capabilities • Comprehensive visualization tools Save this - it will be your competitive advantage when this tool becomes mainstream. 📈 Want the SOP? DM me.
2
2
7
4,985
INSANE N8N AI Agents to Automate ANYTHING! 🤯
7
36
1
376
30,615
Amazon has unveiled Alexa+, a redesigned AI assistant with generative AI capabilities for smarter interactions and deeper personalization. While it offers advanced features like real-time video analysis and email insights, privacy concerns arise over data access and storage. With competition heating up, Amazon must balance innovation with user trust. YouTube link: youtu.be/yXANWfexZT4
1
2
641
Companies are cutting jobs & replacing workers with AI. Efficiency is great, but AI can't replace leadership, creativity, or human connection. The future isn't AI vs. Humans—it's AI + Humans. Thoughts? #AI #FutureOfWork
1
3
695
RAG Intelligent Upgraded: LangGraph + Agentic RAR + Nano-GraphRAG + Claude 3.7 Sonnet (Oxford Univ) Large Language Models Struggle with Sequential Reasoning. For example,if you ask an LLM to solve a multi-step math problem,It might forget earlier steps and give a wrong answer.
2
2
2
1,915