chirdeep

Founder - Covia Labs: Universal Grid for the agent economy. Multi-agent systems need an open, pluggable coordination layer.
AI/ML, Tokenization & data nerd

[chirdeep @chirdeep]

My Substack thing https://t.co/znKhkAS53Z
Let’s do this!!

[chirdeep @chirdeep]

@tarakeeney @HarryStebbings He never responds to my messages. Well done 👏

[chirdeep @chirdeep]

Agentic fabric should be neutral no? Salesforce and serviceNow having full control of your system of record and now even system of action and context seems a bit much. Also, what is your full workflows needs agents from salesforce, serviceNow, SAP, AWS, DataDog, Box? Who runs and owns the Agentic fabric then?

[chirdeep @chirdeep]

@jasonlk Cannot agree more about Agent Fabric - We have been banging on about it.
We call it The Grid for the multi-agent systems. We are building just this. Can we chat?

[chirdeep @chirdeep]

@kenanhsaleh @a16z @speedrun would love to chat ...

[chirdeep @chirdeep]

Relentless delivery @CoviaAI 🚀

[Mike @mikera]

I'm apparently ranked #1 on the GitHub Commits Leaderboard :-) https://t.co/hGpaqLKMwu

[chirdeep @chirdeep]

Before agents become mission-critical, organizations will demand:
– shared execution state
– durable memory
– boundary-aware coordination

That infrastructure comes before dominant agent apps, not after.

If you are exploring multi-agent systems in production, whether as a builder, operator, or partner - let’s talk: @CoviaAI

[chirdeep @chirdeep]

Every major computing shift needed a shared grid before it could scale.

Electricity → power grids
Internet → TCP/IP & HTTP
Cloud → control planes

As agents move from demos to real work, the same pattern is repeating.

Agents don’t just need intelligence.
They need a place to run together.
Full essay ↓
https://t.co/KcKb2dmJ8Q

#AI #MultiAgent #CoviaAI #UniversalGrid #AgentAI

[chirdeep @chirdeep]

Grids succeed because they’re shared.

Power grids don’t belong to machines.
The internet doesn’t belong to apps.

The agent economy won’t be different.

[chirdeep @chirdeep]

As AI agents move from assisting humans to acting inside organisations, something old returns.

Systems of record.

Not for data, for action.

When software starts making decisions with real consequences, intelligence isn’t the constraint anymore. Accountability is.

Why this matters more than better models 👇

Why Systems of Record Return in the Agent Era https://t.co/nlRAefLq6T
#AI #AgenticAI #CoviaAI

[chirdeep @chirdeep]

You need not just unique agentic identities but also unique operation and asset identifiers too no matter where so you can call them. We will need a universal grid for this agent economy. Been at it for a while. Would love to chat /show this in action @levie if this is of interest

[chirdeep @chirdeep]

There’s a lot of discussion right now about context graphs and AI systems of record. Context is important. But in multi-agent systems, execution fails long before representation does.
If agents are going to operate across real systems, we need ordering, authority, and runtime guarantees, not just better memory.
I wrote about where I think the stack is converging.
Would love feedback from folks thinking about context and system-of-record layers.
https://t.co/Ktyz0d2blm
#multiagent #EnterpriseAI #covia_ai

[chirdeep @chirdeep]

@HarryStebbings Not sure @HarryStebbings - it’s harder to get investor attention today unless you are ready $100m ARR. I have raised a few times before but not seen the level of noise to be drowned out. Or maybe I am not in the right place.

[chirdeep @chirdeep]

@atShruti Algo worked 🚀

[chirdeep @chirdeep]

@StephNass VC Map

[chirdeep @chirdeep]

@jefielding Sent what we are building - would love a chat to discuss in depth.

[chirdeep @chirdeep]

🚀

[NodeOps Network @BuildOnNodeOps]

India’s textile floors have been operating in the dark.

By routing live CCTV data through CreateOS, factories are turning basic digital surveillance into an AI powered control surface for their operations.

→ Over 75 days, cotton processing and manufacturing sites across Maharashtra, Tamil Nadu, Telangana, and Gujarat transformed their on site CCTV backup and monitoring system into a fully digital setup running on CreateOS.
→ They backed up more than 50,000 hours of video data (about 125 TB) into a single workspace, so teams could access, analyze, and monitor everything through a simple dashboard.
→ These were MSMEs, not teams with in house data scientists or full time DevOps, working through remote locations, unreliable local storage, and fragmented oversight.

NodeOps is using CreateOS to bring AI driven, real world solutions to industries that never had this kind of tooling.

CreateOS gave these factories a way to see their operations live and act before things went wrong. Textile manufacturing is just the beginning.

Watch the documentary to see how live CCTV data becomes a real time control layer for factories.

[NodeOps Network @BuildOnNodeOps]

India’s textile floors have been operating in the dark.

By routing live CCTV data through CreateOS, factories are turning basic digital surveillance into an AI powered control surface for their operations.

→ Over 75 days, cotton processing and manufacturing sites across Maharashtra, Tamil Nadu, Telangana, and Gujarat transformed their on site CCTV backup and monitoring system into a fully digital setup running on CreateOS.
→ They backed up more than 50,000 hours of video data (about 125 TB) into a single workspace, so teams could access, analyze, and monitor everything through a simple dashboard.
→ These were MSMEs, not teams with in house data scientists or full time DevOps, working through remote locations, unreliable local storage, and fragmented oversight.

NodeOps is using CreateOS to bring AI driven, real world solutions to industries that never had this kind of tooling.

CreateOS gave these factories a way to see their operations live and act before things went wrong. Textile manufacturing is just the beginning.

Watch the documentary to see how live CCTV data becomes a real time control layer for factories.