probatorem

FDE @mach1ai

[Mach 1 @mach1ai]

Minetta Gould, Product Operations Manager at DNSFilter, shares why she loves Mach 1.

@DNSFilter is a leading provider of DNS security, helping organizations block online threats and filter unwanted content.

Watch the full conversation with Minetta and Rob Meinhardt on how DNSFilter uses Mach 1 for customer support: https://t.co/3e5m6kaqT3

[Mach 1 @mach1ai]

We tested five models on the same customer support workflow in Tower, then asked each to update it when the refund policy changed.

All five passed all 10 test cases. The cost to build and update ranged from $0.48 to $15.61.

Full results below. https://t.co/Ekd2NOMCoY

[probatorem @0xProbatorem]

Glad to see this is getting some use in the big ‘26

[Evan Hubinger @EvanHub]

Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.

[Mach 1 @mach1ai]

The operator controls the work. Mach 1 handles the infrastructure underneath it. https://t.co/SGEaKBU8tb

[Mach 1 @mach1ai]

https://t.co/DlkPkPmPvT

[Mach 1 @mach1ai]

Forward Deployed Engineer Sam Ouzounian describes the years of knowledge and experience compiled into our tools.

[Mach 1 @mach1ai]

Happy Fourth of July from Mach 1! 🇺🇸

[Chris Olson @chrisolson]

One thing we’re learning at Mach 1:

Agents don’t just automate work. They create visibility.

Once an agent is inside a workflow, it can spot the patterns a dashboard misses: where customers are confused, where process breaks down, what language is resonating, and what the business should change next.

[Chris Olson @chrisolson]

Sonnet 5: $2/M input, $10/M output.
Sonnet 4.6: $3/$15.
The cost pressure is clearly being heard.

[Mach 1 @mach1ai]

Rate confidence scores, escalate low confidence answers, and improve your agents over time.

[Mach 1 @mach1ai]

One of our customers sells hardware through the channel, where partners get different pricing and terms.

We built a model of how an agent handles that complexity. Here's what it looks like:

[probatorem @0xProbatorem]

Wild

Production-ready agents in 10 minutes, powered by years of accumulated domain expertise.

Sounds like a ticket out of agent PoC purgatory

[Chris Olson @chrisolson]

Mach 1’s latest capabilities in helping operators build and deploy agents into their business.

In this demo, I recreate a sales agent for Trace, the same kind of agent we built three years ago. What once took months can now be built and deployed in under 10 minutes.

Mach 1 brings intelligence to your operations, helping teams build agents that can handle complex, real-world processes.