SoduWork (@SoduWork) header
SoduWork (@SoduWork) profile picture

AI tools, gadgets, games, and the occasional software rant. Useful finds, curious builds. Supervised by cats.

Joined October 2026
SoduWork is now open source! 🚀 An AI-native engine for enterprise workflows: smart dispatch, built-in governance, and verifiable task delivery. Star, fork, or contribute: github.com/mdiri-ai/SoduWork #OpenSource #AI #GitHub
23
5
11
33
9,875
Good work shouldn’t disappear when a task ends. SoduWork helps teams save useful methods and reuse them on future tasks. Because when the water pump breaks on Mars, starting from scratch is a terrible plan. #SoduWork #AIWorkflows
2
Meet Crazy Operations, the content creation app in SoduWork. Start with a topic, explore ideas, and develop a video in different styles—all in one workflow. Even when the newsroom is chasing a rumor. #SoduWork #AIContent
2
One task, different AI roles. SoduWork brings agents together to plan, build, and check their work—while you stay in control of the result. Useful for fixing a workflow. Or a bridge. #SoduWork #AIAgents
3
Code has GitHub. Team know-how deserves one too. Local check → submit → review → publish → grant → improve → revoke. That's how a work method becomes an enterprise asset at SoduWork — with a full audit trail. 87 seconds from our global launch, demo included: ↓
8
What does it take to bring AI agents into real enterprise workflows? Multi-agent collaboration. Built-in governance. Verified delivery. SoduWork — open source, MIT. Repo in replies. #AIAgents #OpenSource
1
15
SoduWork brings different AI foundations into one workspace. Switch the foundation as your work moves forward, without starting from scratch. Think of it as finding another route when the sky is full of flying cars. #SoduWork #AI
1
23
Asking which model to standardize on is the wrong question. Models get swapped every quarter. Your plugins and approval flows don't. Standardize the layer that survives the swap.
1
2
65
Your team's AI moat isn't the model — everyone runs on the same ones. It's what your best people know: the checklists, the review gates, how work actually gets done. SoduWork turns that into reusable Skills agents execute and validate — self-hosted, so it stays yours.
1
1
3
59
One agent can do a task. A team of agents can run an operation. That's the shift behind SoduWork: agents that plan together, hand off work, and sign off delivery against evidence — not a chatbot that stops at "task done." Open source today. Repo in the reply.
1
1
4
90
5 industry leaders. 5 insights. One vision: AI’s future is organizational intelligence — not just smarter agents. 🚀 #AIAgents #EnterpriseAI
1
2
54
Meet SoduWork — an open-source, enterprise-grade AI-native engine. Multi-agent collaboration, built-in governance, verified delivery. MIT licensed. Build with us: github.com/mdiri-ai/SoduWork #OpenSource #AIAgents
3
2
5
11
17,162
AI's next leap isn't just smarter agents. It's better collaboration. Individual AI experiences are valuable, but their real potential emerges when knowledge can be captured, reused, and shared across teams.
1
1
34
SoduWork launches tonight! 🚀 How can enterprises verify AI outputs, clarify accountability & preserve workflows? Join our Nov 16 live stream to explore enterprise AI agents. MIT licensed open source. Come build with us! #SoduWork #AIAgents #OpenSource
3
11
3
144
1,531
🔴 We're live now: SoduWork Global Open-Source Launch From solo AI to team intelligence — enterprise-grade AI-native engine, now open source. ⭐ github.com/mdiri-ai/SoduWork Join the live: live.csdn.net/room/csdnnews/…
3
2
3
5
16,844
Qwen4-Max is here, but its weights aren't open. Qwen has helped make powerful models more accessible. The question now is how much of that progress will remain available to run locally. #AI #Qwen artificialanalysis.ai/models…
1
32
SpaceX caught Starship after a trip to orbit. The booster catch failed, so the full double catch will have to wait. But watching a spacecraft come back from space and get caught by a tower still feels like sci-fi. #SpaceX apnews.com/article/df394c1b5…
1
36
A security researcher suspected a flaw in uBlock Origin. GPT-5.6 helped find it. The bug could let a malicious site run code on other websites—and it's now fixed. AI won't replace security researchers. It just gave them another teammate. #AI iter.ca/post/ubo-uxss/
1
63