Grep.AI is where you create agents to automate repetitive work you can't afford to get wrong. Run them via API, MCP, on a schedule or as a batch.
San Francisco, CA
- Tweets 299
- Following 8
- Followers 379
- Likes 103
We've been shipping agents for a while. Here's what we've learned along the way...
We’ve spent the past three years building agents for real production work. Our architecture changed a lot along the way. Some of those changes worked. Others produced expensive and occasionally painful lessons. Read this to learn from our mistakes.
3
600
A customer asked me last month how other companies are rolling out AI agents. What's working, what it costs them, and where things stall after the pilot.
I answer some version of that question on calls most weeks. I've never written any of it down to share with others, so I'm restarting my Substack newsletter.
I started the original newsletter, The Hitchhiker's Guide to AI, a few weeks after ChatGPT launched, when everyone was still working out what generative AI was. It let me learn about the space as an outsider looking in. Writing about AI every week eventually led me to the conviction to found an AI startup, and the writing took a back seat. That was three years ago.
The newsletter is now back as The Operator's Guide to AI because I'm no longer an outsider looking in, and neither are you. We're all insiders, trying to build, deploy, and scale AI systems, and I want to share as much as I can about what we're learning along the way @grepdotai.
But there are a million different AI publications, so why should you subscribe to mine?
1/ We've been deploying agents into production since 2023, worked with every generation of models, and made plenty of mistakes along the way. I'll be sharing a lot from first-hand experience, not only of deploying agents at large enterprises, but also of how we use AI internally too.
2/ We speak with large enterprises all the time about their AI transformation efforts: no one has really worked it out, and everyone is a lot earlier than you think. I'll be sharing what we're seeing on the ground.
3/ There's a lot of noise in the AI space, with new models launching every week, new capabilities being announced, and endless fundraising announcements. I want to cut through the noise and focus on what practically matters for folks deploying AI and agents at scale.
If that sounds interesting, you can subscribe below...
The first post on why agents fail in production goes out this week.
4
2
19
493
We’ve been busy!
In the past few weeks we @grepdotai have launched:
- Make-your-own managed agent
- Loops and monitors
- Media agents
- Our MCP
- A bunch of UX updates and more.
I’m recapping what we’ve shipped in a demo. Showing y’all how I built @brieflyin using our managed agents.
2
4
327
GREP.AI (YC F26) retweeted
In the past few weeks we @grepdotai have launched:
- Make-your-own managed agent
- Loops and monitors
- Media agents
- Our MCP
- A bunch of UX updates and more.
I’m recapping what we’ve shipped in a demo. Showing y’all how I built @brieflyin using our managed agents.
5
6
20
5,011
GREP.AI (YC F26) retweeted
Just added a escalation tool to our @grepdotai cloud agents. I saw a Tweet about this earlier this week and built it. With the right controls in place, this will be super useful.
1
4
357
GREP.AI (YC F26) retweeted
We released custom agents in @grepdotai this week. I have been using them for long-running tasks. In a couple of hours, Fable used them (through our MCP and @cloudflare's ) to build briefly.in. News + podcasts on multiple topics: AGI, San Francisco, US News, etc. Check it out.
1
6
11
727,732
Custom agents are live in grep.ai! Check them out.
We released custom agents in @grepdotai this week. I have been using them for long-running tasks. In a couple of hours, Fable used them (through our MCP and @cloudflare's ) to build briefly.in. News + podcasts on multiple topics: AGI, San Francisco, US News, etc. Check it out.
5
3
17
503,690
At @ParchaHQ we saw coding costs explode so we built our own IDE that does routing at the harness level e.g. Claude Code for planning, Codex for long running execution, Cursor for bugs. Plus custom routing, remote sessions, a browser inspector and voice interface.
Now it's open-source! 👇
7
9
34
8,254
Awesome to see @gilbert sharing how they use @Grepdotai for episode research on @AcquiredFM in a GQ article this week.
There are so many non-obvious use cases for doing deep research on a business, and Grep is the best at it.
1
1
7
248
GREP.AI (YC F26) retweeted
It was a big magical in the onboarding when @grepdotai researched and started to set itself up.
When you use an AI tool for the first time, it doesn't understand you or your work. We're changing that with the latest version of @Grepdotai.
At onboarding, with your opt-in, Grep uses its deep research to build a profile about you that it then remembers for future interactions.
3
4
11
2,065
When you use an AI tool for the first time, it doesn't understand you or your work. We're changing that with the latest version of @Grepdotai.
At onboarding, with your opt-in, Grep uses its deep research to build a profile about you that it then remembers for future interactions.
4
10
3,160
Deepfake fraud losses hit $12.3 billion in 2023. Deloitte projects $40 billion by 2027.
Deepfake attempts surged 2,137% in three years. Only 7% of anti-fraud professionals say their org is even moderately prepared.
We ran the full risk analysis across 7 domains.
1
1
282
The detection gap is where the fraud lives.
Lab accuracy: 94-96%
Real-world accuracy: 65%
Human detection rate: 24.5%
42.5% of all financial fraud attempts are now AI-driven. 85% of orgs experienced at least one deepfake incident in the past year.
Insurers quietly excluded AI-generated fraud from standard cyber policies as of January 2026.
1
38
North Korea generated $2.8B using deepfake IT worker infiltration across 40 countries.
Regulators are moving. FinCEN issued its first deepfake-specific alert. EU AI Act penalties hit 7% of global turnover.
Full intelligence brief:
grep.ai/research/deepfake-fr…
Run research like this yourself:
grep.ai
51
The Strait of Hormuz has been blockaded for 65 consecutive days.
- 4.6x the longest closure on record.
- 20% of global oil supply blocked.
- Brent at $112. Physical crude at $132.
And the real shock has not hit yet.
1
4
174
Country exposure:
Japan: 87% of oil imports via Hormuz. 230 days of reserves burned to 165.
South Korea: 80% dependent. Fewer reserve days. No pipeline alternative.
India: 65% dependent. 10 days of strategic reserves. 𝗧𝗲𝗻.
Basel III stress tests cap Hormuz disruptions at 30 days. We passed that 35 days ago.
1
53
Full financial risk intelligence brief available with all eight risk vectors mapped, asset class impact matrix, CRO repricing checklist, Day 1 vs Day 30 vs Day 65 comparison.
Link - grep.ai/research/strait-of-h…
18