惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

V
Vulnerabilities – Threatpost
V
Visual Studio Blog
博客园_首页
Last Week in AI
Last Week in AI
J
Java Code Geeks
V
V2EX
博客园 - Franky
Apple Machine Learning Research
Apple Machine Learning Research
云风的 BLOG
云风的 BLOG
博客园 - 司徒正美
N
Netflix TechBlog - Medium
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
爱范儿
爱范儿
B
Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
U
Unit 42
W
WeLiveSecurity
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
A
Arctic Wolf
S
Security Affairs
博客园 - 聂微东
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
Microsoft Azure Blog
Microsoft Azure Blog
雷峰网
雷峰网
L
LangChain Blog
大猫的无限游戏
大猫的无限游戏
罗磊的独立博客
I
Intezer
有赞技术团队
有赞技术团队
Forbes - Security
Forbes - Security
Engineering at Meta
Engineering at Meta
D
Docker
C
Check Point Blog
N
News | PayPal Newsroom
H
Hacker News: Front Page
T
The Blog of Author Tim Ferriss
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
C
CXSECURITY Database RSS Feed - CXSecurity.com
SecWiki News
SecWiki News
The Last Watchdog
The Last Watchdog
Recorded Future
Recorded Future
量子位
NISL@THU
NISL@THU
T
Tailwind CSS Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
MongoDB | Blog
MongoDB | Blog
MyScale Blog
MyScale Blog
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
I
InfoQ
Hacker News: Ask HN
Hacker News: Ask HN

Mozilla.ai

Open Models are ready for agents. Their APIs are not. Using Octonous as an AI Safety Engineer The Control Layer: Why the Next Era of AI Is About Infrastructure, Not Just Models Introducing Otari: The Open-Source LLM Control Plane Announcing transcribe.cpp Using Octonous as a Product Manager Image Classification Comes to encoderfile What is an LLM control plane? Use the Otari Gateway with OpenCode Otari: Own Your AI Stack | AI Gateway & Hosted Platform AI Got Expensive. Now What? | Mozilla.ai cq exchange: Agents without Borders The Interface Is No Longer the Product VIBE✓: First Defense for cq (Stack Overflow for Agents) Sovereign AI: Control, Choice, and Beyond Geopolitics Encoderfile’s New Format: Why a “Dull” Design Wins The Real Challenge Behind Small Trade Businesses Hardening Your LLM Dependency Supply Chain cq: Stack Overflow for Agents cq: Stack Overflow for Agents llamafile Reloaded: What’s New in v0.10.0 When Shipping Software Becomes Too Easy Mozilla.ai Joins Flower Hub as Launch Partner Owning Code in the Age of AI The Star Chamber: Multi-LLM Consensus for Code Quality
Octonous Open Beta: What We've Learned and Where We're Going
David de la Iglesia Castro · 2026-05-08 · via Mozilla.ai

The Octonous open beta is live. Learn what we discovered during closed beta, the workflow patterns users kept returning to, and the biggest improvements shipped since launch.

Octonous Open Beta: What We've Learned and Where We're Going

From Closed to Open Beta

A few months ago, we introduced Octonous to a small group of early users. We wanted to test a core hypothesis: could an AI assistant genuinely reduce busywork across connected apps without feeling like just another tool to manage? 

The feedback we received shaped almost everything we've built since.

Today, we're releasing Octonous into the wild.

Octonous is now available in open beta to everyone at octonous.com.

Video thumbnail

What We Learned

Running a closed beta taught us a lot about where workplace friction really lives.

We learned that the “blank canvas” of an AI assistant can be daunting. Because Octonous can be used for so many things, it wasn't always clear to users what the best first workflow was for their situation. This ambiguity made it harder to build the habit of using it.

Octonous in-person workshop in Berlin (April 21, 2026)

We also learned where people kept coming back:

  • “Start in chat, then automate" pattern: Users loved drafting and refining workflows through their conversations with the AI assistant.
  • Complete Transparency: See every step of an automation built trust. 
  • Approval Flows: Users loved that nothing gets sent, posted, or updated without their explicit sign-off.

Ultimately, our users wanted more control over what the agent could see and do, clearer logs of what had happened, and better ways to refine a workflow before committing to automating it.

Octonous workshop in Lisbon (March 12, 2026)

We also heard consistent feedback about personalization. Users wanted Octonous to feel less generic over time. They wanted it to remember their preferences, adapt to how they work, and stop asking for context they'd already given. 

And we heard from teams with specific infrastructure requirements who wanted to choose which AI model powers their workflows, rather than being locked into a single provider.

What's New Since Closed Beta

Octonous in-person workshop presentation in Lisbon (March 12, 2026)

Since closed beta, we've shipped a meaningful set of improvements, and we’re excited to share the most impactful ones:

Expanded  Integrations and Triggers

We've significantly grown the list of connected apps, adding Hubspot, Notion, Linear, GitHub, Google Workspace, Salesforce, Typeform, and more.

In addition to running via schedule, automations can now be kicked off by a wider range of events, from new emails and Slack messages to form submissions and calendar events.

Choice-first Architecture

You have fine-grained control over the access scope of the integrations you connect, ensuring Octonous only performs the actions you authorize.

In addition, Octonous allows you to choose among all three major LLM providers: Anthropic, Google Gemini, and OpenAI. And if you have specific requirements or want to use a model we don't offer out of the box, you can bring your own.

Bring your model to feature on Octonous

Refined Human-in-the-Loop

The approval flow is faster and clearer, so staying in control doesn't mean slowing down. You can manually edit the contents of any write action before Octonous executes it.

Streamlined Onboarding

We've added example workflows and starting points to help new users find their first automation faster. You can even ask Octonous for ideas, and it will recommend personalized tasks for you to try.

Memory 

Octonous can now remember details about you and your preferences, so it behaves the way you work. Tell it once that you prefer bullet-point summaries, or that you always post updates to a specific channel, and it will remember. The more you use it, the more it adapts to you.

What Open Beta Means

Open beta means Octonous is available to anyone right now, no waitlist, no invite needed. That said, it's still evolving. If something doesn't work the way you'd expect, use the feedback button in the app. We read every single submission.

Connect the apps you already work with on Octonous

What's Coming Next

We're currently  building team collaboration features so automations can be shared and managed across a workspace, not just by individual users. We're also adding more integrations and upgrading how the agent handles complex, multi-step workflows.

The goal remains the same: less juggling between apps, more time on work that actually matters.

Try It!

Get started for free at octonous.com. Receive 1000 credits when you sign up, connect your apps, and run your first automation!