Giving ChatGPT full access to my Twitter account: should I be worried?
@datachaz
-
Gumloop launches Agent Browsers for AI agents
By
–
NO MCP? NO API? YOUR AGENT STILL HAS A WAY IN@gumloop just launched Agent Browsers for the work that lives behind a browser interface.
— Charly Wargnier ☄️ (@DataChaz) 6 octobre 2026
> secure credential storage and 1Password integration handle credentials
> session replays let you inspect the agent’s actions
> stealth,… https://t.co/Z89o2DS48q pic.twitter.com/4PB0XkwGXRNO MCP? NO API? YOUR AGENT STILL HAS A WAY IN @gumloop just launched Agent Browsers for the work that lives behind a browser interface. > secure credential storage and 1Password integration handle credentials
> session replays let you inspect the agent’s actions
> stealth, -
Clarifying agent activity at scale across an org
By
–
The maths checks out 😅 but it’s not really 166 agents running per engineer simultaneously. The 1M figure reflects agent activity at scale across the org 🙂
-
OvermindLab turns agent traces into training and eval datasets
By
–
YOUR AGENT’S PRODUCTION TRACES ARE NOW TRAINING MATERIAL
— Charly Wargnier ☄️ (@DataChaz) 6 octobre 2026
So cool!@OvermindLab now connects your code and traces, builds a context graph, and turns real interactions into training and eval datasets.
Use them to compare prompts and models, investigate failures, and fine-tune a… https://t.co/Ajh1EpgLDb pic.twitter.com/mNNNYx6wOQYOUR AGENT’S PRODUCTION TRACES ARE NOW TRAINING MATERIAL So cool! @OvermindLab now connects your code and traces, builds a context graph, and turns real interactions into training and eval datasets. Use them to compare prompts and models, investigate failures, and fine-tune a
-
V4 Turbo improves AI agent voice quality
By
–
V4 TURBO FINALLY MADE OUR AGENTS SOUND RIGHT 🤘
— Charly Wargnier ☄️ (@DataChaz) 6 octobre 2026
Stoked for ElevenAgents Architect to keep building ↓ https://t.co/GC0jj47IoB pic.twitter.com/m6zVrWBFxwV4 TURBO FINALLY MADE OUR AGENTS SOUND RIGHT 🤘 Stoked for ElevenAgents Architect to keep building ↓
-
94% of engineering teams adopted AI, only 6% scale AI-Native SDLC
By
–
Last week, I tuned into the @Atlassian State of AI SDLC Digital Summit and something stopped me: 94% of engineering teams have adopted AI. Only 6% have the systems in place to scale a truly AI-Native SDLC. Mike Cannon-Brookes (CEO + Co-Founder at Atlassian) sat down with
-
Agentic AI adoption in healthcare welcomed
By
–
Agentic AI is coming for healthcare and I'm here for it 👏👏👏
-
Agentic AI platform guides patients through full care journey
By
–
AGENTIC AI IS COMING FOR HEALTHCARE TOO
— Charly Wargnier ☄️ (@DataChaz) 5 octobre 2026
The big shift: @nollahealth isn’t another medical chatbot.
It can guide a patient through the actual care journey.
→ intake
→ assessment
→ treatment plan
→ prescribing when appropriate
→ follow-up
For now, it’s launching with acne… https://t.co/jSpjQAI4NFAGENTIC AI IS COMING FOR HEALTHCARE TOO The big shift: @nollahealth isn’t another medical chatbot. It can guide a patient through the actual care journey. → intake
→ assessment
→ treatment plan
→ prescribing when appropriate
→ follow-up For now, it’s launching with acne -
Photon raises $4.5M for AI agent messaging framework
By
–
MUSE AND INSTINCT GET THE SPOTLIGHT. PHOTON POWERS THE MESSAGING BEHIND THEM.@PhotonHQ just raised a $4.5M seed round.
— Charly Wargnier ☄️ (@DataChaz) 5 octobre 2026
Its open-source framework, Photon , connects AI agents to iMessage, WhatsApp, Telegram and more.
Some crazy traction since April!
→ 450K monthly npm… https://t.co/l1qXggcJfvMUSE AND INSTINCT GET THE SPOTLIGHT. PHOTON POWERS THE MESSAGING BEHIND THEM. @PhotonHQ just raised a $4.5M seed round. Its open-source framework, Photon , connects AI agents to iMessage, WhatsApp, Telegram and more. Some crazy traction since April! → 450K monthly npm
-
Space Bunny Alpha stealth model tops OpenCode usage
By
–
THIS STEALTH MODEL IS UBER FAST ⚡
— Charly Wargnier ☄️ (@DataChaz) 4 octobre 2026
It’s also been the #1 most-used model on @OpenCode over the past five days.
Space Bunny Alpha combines flash-level responsiveness with:
> a 1M-token context window
> strong coding capabilities
> multimodal input
> adjustable reasoning
But… pic.twitter.com/GZRITnl7OxTHIS STEALTH MODEL IS UBER FAST ⚡ It’s also been the #1 most-used model on @OpenCode over the past five days. Space Bunny Alpha combines flash-level responsiveness with: > a 1M-token context window
> strong coding capabilities
> multimodal input
> adjustable reasoning But