What you are citing appears to me as a business use case of classical machine learning models not necessarily LLMs, and demand forecasting. Frontier AI is clearly not dependent on such data, and has many outside data sources and I'm even sure sensitive AWS data would be of good use as training data, not to mention the legal/contractual obligations AWS has in protecting that data, what am I missing?

That was just an example anecdote to highlight the culture around respecting privacy and security. Obviously AWS would never allow for training on customer data, but even for metadata or data of our own internal systems, things are super strict.

Public datasets that are liberal with licensed access is one thing, but the best frontier labs clearly went well beyond that.

I'm curious if you have a few mins for feedback, what are the top 2 things here that omnigent does that is significantly better for you than latest cc/codex which can launch subagents, auto save memory of a project.

I'm wondering that as these core tools continue to enhance and add these capabilities, how much of a benefit these meta harnesses actually provide.

I agree the gap is narrowing quickly, especially with workflows in CC. For me, a few large advantages still remain:

1/ Allowing me to easily plug in any harness, using any provider, and make it a first-class worker. Omnigent has out of the box ACP support and it's trivial to use that to add first-class support for any harness out there. I love the ability to have CC + Opus plan, Codex + Luna implement, Pi + Qwen 3.8 give a tie-breaking opinion on a design decision that Opus flagged and Grok and Codex couldn't agree on, all orchestrated by a model of my choosing from any provider using Omnigent's main agent harness.

2/ Reusable agent systems rather than just reusable workflows. You can define agents in YAML whose subagents embody particular roles, with different models/harnesses, skills, plugins, tools, etc. preconfigured for each one.

Of course, claude workflows are now durable but Omnigent's agnt definitions are a bit more abstract in that they define the subagents that are available and how they should work by default rather than the workflow itself (i.e. the specific JTBD). If I have a common workflow that consists of, for example, Sol + Codex writing some script to scrape some data, Pi + a cheap DeepSeek-tier model formatting that data en masse, then Fable + CC doing some advanced analysis on it, I can embody that with a yaml agent definition that I can then use to run with my task of the day as a prompt. All of this is orchestrated by a model of my choice using Omnigent's harness.

This might look like: 'smart scraping agent with all sorts of scraping skills and tools pre-loaded', a 'bulk data processing agent with a cheap, fast model and plenty of pandas/numpy skills preloaded', and 'frontier model to interpret and reason on the implications of the processed data'. The main agent would have instructions about the general workflow of such tasks and when to invoke and delegate tasks to which subagent. The definition describes the workers available to the orchestrator and how they should generally behave, rather than hard-coding the workflow itself. I love that I can create those definitions and re-use them.

All that said, I am sure the labs will come up with their own similar products to (2) (e.g. dots today). I also recently noticed that Claude Code now has subagent 'teams' rather than just 'general-purpose'/'explore' subagents, and these seem to be longer-lived. This seems to be encroaching on the agent yaml definitions, albeit with less fine-grained control on my end. Therefore, the tl;dr (for me at least) is vendor neutrality; I don't think we'll ever see a product coming out of a frontier lab that eagerly delegates a task to their competitor's model (and bank account).

Could this work to drive robots in a confined space without humans, and time isn't a huge factor, where full automation with scale can still be economical, like in a lights out environment?

This is by far the most annoying thing about Amazon to me. They clearly are doing a disservice to their consumer with this. Many times I go straight to google to search for Amazon products, which leads to better results

No one does it like Amazon, it requires users to walk through the sponsored ads before they can see organic search results, almost no other platform does this online. Walmart has a single strip of sponsored results at top, immediately leading to organic results, others like Target, HomeDepot do not mix in sponsored results with search results.

B&M is different, Ads are there, but you aren't forced to look at them like Amazon does.

> B&M is different, Ads are there, but you aren't forced to look at them like Amazon does.

Ignorance is bliss. Unfortunately, even your physical Walmart store is also selling a lot of shelf space to manufacturers/advertisers (similar to how Amazon sell the first slot, or at this point first 'page'), especially those checkout lines (https://www.cbsnews.com/news/how-grocers-wring-extra-cash-ou...). That's why your on-boots retail staff can't change those planograms (which defines how to layout shelf space in a grocery store) willy-nilly, they're designed by the highest bidder. That's why those soda or chips seems placed in too convenient slots that are also 'coincidental' to those currently advertised in other ways, because it's, plain and simply, paid to be there.

Walmart is not making me move things(ads) out of the way to proceed to checkout, I don't know how much simpler I can get beyond this.

Everything comes with a risk, and people usually take it with a decent understanding of how to mitigate it mostly.

If we start thinking along the lines of technology has risk and we shouldn't use it, we should go back all the way to the discovery of fire as we all know fire can cause a lot of damage if in the wrong hands.

Technology can make lives safer. However, software is never secure, cannot be made secure, this is empirically proven to be a 100% valid position.

In as software is not secure, and can not be made secure, using it for important records storage makes zero sense, unless you a) have full backups offline in physical, non-digital, read only medium or b) just don't do it online, at all.

And my point is, it's not worth the convenience.

But physical records also cannot be made secure. And the convenience is gigantic. I think that it is worth it, as clearly does much of the world.

Read the scope I discussed. Physical records can be made exceptionally secure. They can be secure on location (archival location, with guards). They also aren't reachable by every human being on the planet who wants to infiltrate.

Think about it. To destroy the public archives, of which there is typically more than one, you must travel to said location, breach it with weapons and other means, and destroy it. Or, you can sit in your parent's basement in your pajama's, and hack and destroy.

There is not even remotely the same risk profile.

And if you think something is "good" because 'the world thinks it is good', then I have to ask you why you're validating something via popularity. You know what else was popular? Fossil fuels. Smoking. Using uranium in makeup for women. Something being accepted, and being fun or convenient, doesn't make it correct, sensible, or right.

So please describe the horrible and incredible "gigantic convenience". Because I lived before the internet, and now after, and yes it is convenient.

But it certainly isn't a 'gigantic' one, nor is it sensible compared to the insane attack surface and risk.