ruben

Orchestrating more agents to keep up on AI. Tinkering @tenopsai

[ruben @rubentenops]

@wordgrammer I have that but added my own RAG system doing simple hybrid search and just dumping the top 10 results and letting agents go from there deciding what else to look up.

And to optimised retrieval surface I have a scheduled agent going over new docs and adding/updating front-matter

[ruben @rubentenops]

@lukepierceops Playbook

[Luke Pierce @lukepierceops]

Late 2024: AI agencies sold n8n workflows for $2-$5K.

Mid 2025: they pivoted to AI agents for $5-15K.

Today: Claude Code ships in hours what used to take weeks, and most agencies are still pitching 2024's playbook.

I spent 2 months rewriting mine for where we actually are in April 2026.

Inside:

→ The offer closing $25K-$60K projects right now
→ Top 5 industries worth selling to this quarter
→ Content schedule generating my inbound (exact post types + cadence)
→ LinkedIn + cold email sequences booking calls today → My 4-call sales process from first touch to signed
→ The strategy doc + proposal template I'm using to close
→ 3 live client builds my team is shipping this quarter

BONUS: First 100 people also get 2 discovery call recordings from my own sales process.

Like + RT + reply PLAYBOOK and I'll DM you the link. Make sure to follow me so I can DM you.

[ruben @rubentenops]

@tautologer Seems to be a first time experience for everyone 😂

[ruben @rubentenops]

@neural_avb Have you tried any of the new age search APIs?

@ExaAILabs @p0 @tavilyai

They have paid plans but a lot of free credits on signup + I use them only in addition to fire crawl and my own proxy setup.

But it’s great for a research skill that uses all of those and synthesize

[ruben @rubentenops]

@dan__rosenthal OUTBOUND

[ruben @rubentenops]

@moneyfetishist Appreciate it, this is very insightful.

The consultant channel is already on my radar and the long sales cycles as well. Scary but like you say, there's a moat in just persevering and doing the hards things others skip

[ruben @rubentenops]

@Aidanb2b MAGNETS

[ruben @rubentenops]

With “permissioned inference” you mean when platforms deploy custom fine tuned models for a customer?

Custom fine tune model will definitely be required but they are expensive.

Especially when we’re just talking about context graph and decision trace retrieval, the custom models matters less

[ruben @rubentenops]

@scaling_shields VAULT

[Andrej Karpathy @karpathy]

New supply chain attack this time for npm axios, the most popular HTTP client library with 300M weekly downloads.

Scanning my system I found a use imported from googleworkspace/cli from a few days ago when I was experimenting with gmail/gcal cli. The installed version (luckily) resolved to an unaffected 1.13.5, but the project dependency is not pinned, meaning that if I did this earlier today the code would have resolved to latest and I'd be pwned.

It's possible to personally defend against these to some extent with local settings e.g. release-age constraints, or containers or etc, but I think ultimately the defaults of package management projects (pip, npm etc) have to change so that a single infection (usually luckily fairly temporary in nature due to security scanning) does not spread through users at random and at scale via unpinned dependencies.

More comprehensive article:
https://t.co/EJAZbqAPIQ

[ruben @rubentenops]

@goodhunt @DSPyOSS @Thom_Wolf Can agree with that.

I’m new to dspy myself since about a week and it shifted my thinking quite a lot about what works right now.

For reference 3 wees ago I started playing around with auto research with excitement but after a week ran into so many obstacles

[ruben @rubentenops]

@LoganTGott Funnel

[ruben @rubentenops]

@boringlocalseo Local rank

[ruben @rubentenops]

@jspujji GA

[ruben @rubentenops]

@sudiprokaya @laminalabs @ycombinator Been doing the same with a friend. Just Claude codex and our personal harnesses and compound systems.

Building something around context graphs and decision trees

[shirish @shiri_shh]

Farmer pays $5–$8 per cow per month.

A New Zealand company puts a solar-powered smart collar on cows.

It tracks location 24/7, health, temperature, chewing activity, breeding.

Farmer just opens a simple app and draws a line on the map.
That line becomes the fence.

As cows approach the boundary, the collar beeps and vibrates.

With one tap, the whole herd moves to fresh grass or the milking shed.

No physical fences. Less labor. Huge cost savings for farmer.

Already on 700k cows across New Zealand, Australia, and the US.

and now in talks to raise at a $2B valuation led by Peter Thiel.

[gabriel @gabriel1]

start a company assuming:
1) models will become 10x better
2) the only bottleneck for humans is making as many well informed decisions as fast as possible in a great interface

lovable are so impressive, they understood this at gpt 3.5 and the same logic still holds

[ruben @rubentenops]

@mstockton What would you recommend the 25 year old guys do?

The obvious answer would be find experts to partner with.

But what would be the pitch? What are the veterans looking for?