Brad

Helping founders ship MVPs fast 👇

[Brad @BradAI]

Coming to Australia soon.

[Kevin Bass @kevinnbass]

The truth is called hateful in Belgium. Therefore speaking it is against the law.

This is more frightening than calling the truth "misinformation". They have stopped pretending completely. They are openly admitting that they have criminalized the truth.

[Brad @BradAI]

Wow

[Teknium 🪽 @Teknium]

I’m sorry that you’re this desperate that you will take such an unscientific benchmark instead of any established one.

Also qwen local is one of the most random length models there is with all its looping. And we smoke you all on quality benchmarks on every open model. Here’s wildclawbench by internlm, same speed on open models, much better results.

[Teknium 🪽 @Teknium]

Will see if we can get this into Hermes Agent today!

[Brad @BradAI]

A big difference between Hermes and OpenClaw.

Hermes add what the people ask for.

OpenClaw adds a bunch of random shit that breaks stability of the agent.

[Shubham Saboo @Saboo_Shubham_]

This is insane 🤯

Hermes Agent team (@NousResearch) is going from X comments to fully working features in hours.

10x Builder is a thing of past, 100x builder is now real with Agents.

[Brad @BradAI]

Add in a couple Cursor agents, a few Claudes and a Hermes and now we're talking.

[jinjingliang @JinjingLiang]

Me watching Codex do my work

[Teknium 🪽 @Teknium]

We just hit number one globally across all AI apps on OpenRouter.

Super grateful to the nearly 1000 contributors who've helped make Hermes Agent great, thank you!

What do you want to see next?

[Brad @BradAI]

Voices are great, latency needs work.

Even through browser latency is too high for successful calls.

To put it into context our calls routed through telephony using various providers has less latency than this.

I'd love to add xAI into our stack as we are tech agnostic. Just needs improvement for production calls.

[Teknium 🪽 @Teknium]

Introducing Hermes Agent v 0.13.0

- Multi-Agent orchestration through the Kanban system
- Enforced goal completion with /goal
- Big optimizations for disk usage
- Much more extensibility, custom LLM Providers, custom gateway channels, and much more

[Brad @BradAI]

You're overcomplicating you're AI workflows.

[Brad @BradAI]

Beta seats open: https://t.co/bUX1SGvOxw

[Brad @BradAI]

Most productivity advice breaks the moment you add young kids, client deadlines, and a real household.

So I’m testing something narrower:

A 14-day operating system sprint for agency and consulting founders with young kids.

Not “wake up at 5am and optimize your life.”

More like:
- protect one real work block
- create one repeatable family ritual
- fix one household handoff
- map next week before it owns you

[Brad @BradAI]

@enjojoyy What was the reason you went this route?

I have one that runs multiple businesses, projects and personal without any issues. Separated by TG Topics.

Wanting to understand if I am missing some big unlock.

[Brad @BradAI]

Actual useful update.

[Teknium 🪽 @Teknium]

Traditional cron jobs are great for silent tasks on a machine, and Hermes Agent cronjobs are great for extending that to your agent, but why not utilize the gateway and hermes' cron to access things that don't need to cost an agent's time across any messenger service you have connected?

Just `hermes update` and ask your agent to setup cronjobs that need no agent in the loop, like running a systems diagnostic script that reports info to you every 12 hours, pulls in an RSS feed, and send it over to you on telegram, whatever you can think of, while saving a lot of money on purely programmatic tasks!

If the script has no output, you wont get a ping either

PR: https://t.co/VumGUiLTnQ

[Brad @BradAI]

All these people saying their team of agents codes 24/7 while they’re sleeping…

Honestly, what are you building?

I use one agent and can bang out an entire app in a couple of hours.

[Brad @BradAI]

We already placed our chips.

Every person remaining will be paired with a specialist agent that performs the work of 10 people.

Welcome to the future.

[Brad @BradAI]

Yesterday I sat down with a couple of friends in my building and gave them a very basic AI demo.

Not a polished keynote.

Not a “watch me use 47 tools” performance.

Just normal prompting, in real time, while we were talking.

And it completely changed how they saw what’s possible.

[Brad @BradAI]

@bradmillscan Fellow Brad, stop torturing yourself. Make the switch. I promise you'll be glad you did.

[Brad @BradAI]

@Teknium I can confirm it works.

Well done.

[Brad @BradAI]

This is how I go from booked call to a detailed proposal for $2,500+ AI agents plus monthly retainers in about 20 minutes.

No complicated stack.

STEP 1
Before the call, I give my agent the participants, company, and purpose of the meeting. Then I ask it to map the call and prepare the questions I need to ask.

STEP 2
During the call, I use Google Meet to record and transcribe. That’s enough. I keep the AI-generated question map open beside the meeting window so the conversation stays focused.

The call is mostly discovery.

I’m trying to extract:

- pain points
- manual workflows
- time wasted
- missed revenue
- customer friction
- what the problem is costing them
- what a successful outcome would be worth

STEP 3
After the call, I drop the transcript into my agent and have it create the proposal.

It includes:

- what we heard
- their pain points
- the AI agent/workflow solution
- how it solves each problem
- examples or case studies
- pricing
- next steps

STEP 4
Then I have it apply everything to my branded PPTX proposal template.

Optional but useful: upload the proposal to Canva so you can track engagement.

Did they open it?
Did they share it?
Which pages did they spend time on?
Was it pricing?
The solution?
The implementation plan?

Now your follow-up is based on behaviour instead of guessing.

Most people send generic proposals and follow up with “just checking in.”

AI lets you send something specific, polished, and directly tied to what the prospect told you on the call.

That’s the difference.

Specific beats polished.

Specific and polished closes.