Zach Henderson retweeted
This is one of my favorite quotes that I've ever read. Behind it is the TRUTH that explains why every time Christianity seemed "doomed," it never came to pass.
I love this so much.
Zach Henderson retweeted
We're clothed in the righteousness of Christ✌️✌️, we should boast in nothing other than the works of Christ 💯💯.
Zach Henderson retweeted
Replying to @patriotic_em88
Ultimately, I came to realize He's with me even when I'm not with Him. Every thought, every word, every moment is shared with Him.
While you're "trailing off," He is carrying you.
The "I'm sorry" turns into "thank you."
Zach Henderson retweeted
I LOVE a good story about faithful courageous Christians. William Tyndale’s story puts steel in my spine. Wait for the payoff at the end where God answers his *final* prayer…
Zach Henderson retweeted
The Bible Doesn’t Tell Us to Be Tolerant — or to “Not Judge.”
Modern culture calls silence “kindness.”
But Scripture calls it cowardice.
The Bible never commands tolerance of sin - it commands truth:
“Love must be sincere. Hate what is evil.” - Romans 12:9
“Expose the deeds of darkness.”
- Ephesians 5:11
Even Christ rebuked the Church in Thyatira for being too tolerant:
“You tolerate that woman Jezebel… by her teaching she misleads My servants.” - Revelation 2:20
And “judge not” doesn’t mean stay silent. It means don’t judge hypocritically.
Jesus said:
“Judge with right judgment.” - John 7:24
Love calls sinners to repentance.
Tolerance leaves them comfortable in sin.
Love says: “Turn back to God.”
Tolerance says: “Stay as you are.”
Christ loved people deeply, but He never tolerated sin.
“Go and sin no more.” - John 8:11
The world says: “Don’t judge.”
The Bible says: “Stand firm.”
“Come out from among them and be separate.” - 2 Corinthians 6:17
The world wants silence.
Christ calls for courage.
True Christians don’t tolerate sin - they confront it with truth, for the sake of the soul.
That isn’t judgmental.
That’s biblical. ✝
Zach Henderson retweeted
One of the strongest speeches I’ve heard in a while
I will never skip a Pastor Philip Anthony Mitchell video 🙏🏼
Zach Henderson retweeted
Just got off the phone to my insider…
He said, “Fear not, for I am with you.”
The last thing he said was…
“I am the way, the truth, and the life.”
I just thought I’d pass on the message.
Zach Henderson retweeted
🚨 There is a seemingly meaningless detail in the Genesis story of Adam and Eve that almost everyone skips right over.
@CarlosErazoGTG pulls at the thread and it result blows Pastor @howertonjosh's mind 🤯
Zach Henderson retweeted
I think about this a lot.
The U.S. Navy has operated nuclear-powered aircraft carriers and nuclear-powered submarines.
Currently, there are 11 big-deck nuclear-powered aircraft carriers (10 of which are Nimitz class) and 71 submarines in service. And they are primarily operated by 18–30-year-old men.
However, climate alarmists and an entire generation of people inflicted with Jane Fonda's China Syndrome will try to convince you that nuclear fission is too dangerous for society to use for baseload power.
We are a deeply unserious society.
Zach Henderson retweeted
There's a huge opportunity right now in being the deployment layer for AI into the economy. The amount of work it takes to change out workflows in enterprises tends to be far greater than anyone realizes or would prefer. Clearly this is what the applied layer of AI is going to look like in the form of software and agents, but also it opens up new services firms opportunities.
Legacy systems need to be moved to the cloud, data organization and access needs to be updated, software needs to be connected to agents in new ways, workflows need to be reengineered for agents, HITL needs to be figured out for the process, evals need to be generated and maintained, and the entire system needs to be continually updated as new models get released and new capabilities emerge. And the full list may even be longer.
AI is not the same as just deploying software. Software you generally did the implementation of an existing, well understood category of technology, then stepped back and the customer kept running. With AI agents, you're delivering actual work augmentation to the organization, which has a completely different set of complexities associated with it. You're no longer deploying tools that the company is merely enabled by, you're deploying work output in a process. Completely different implementation and enablement process.
As a result, this is going to open up lots of new kinds of firms and plays for existing firms to diffuse AI into organizations. We're going to see approaches by industry, by size of company, and by problem inside of companies. Traditional SIs will modernize and adapt (some will clearly not adapt as well), and new entrants will also be founded in this period that take advantage of this window. Great time to be an FDE or FDE firm.
Box CEO Aaron Levie (@levie) calls out the MASSIVE opportunity for AI deployment services:
"every single one of those companies, whether that's a 50-person firm or a multi-100,000 person firm, is gonna need an army of people to go in and help them with that transformation"
"when you go to that law firm and you go to that pharma company and you go to that bank, they need something that bridges the core technology to their workflow in their business process"
"somebody has to go into that organization and get it set up, and somebody has to go and provide domain expertise to this model so it really understands our particular business process"
Vendor FDEs are incentivized to get you hooked on their platform and to spend more money.
Indie FDEs are incentivized to use the best tool for the job and to save you money.
Hire indie FDEs.
anyone thats worked at mid market and enterprise knows this is the only way
I've spent the past few years running AI strategy or implementations (ibm,sli , and independently) for various teams
this involved deploying exactly what aaron's explaining (hybrid cloud, ai POCs)
it's very hard to sell ai software off the shelf thats self serve in enterprise, so you need to have service backed by a platform
i.e say you're working with insurance company that wants to become ai native in underwriting
I would map out the entire underwriting process, look where we can actually deploy agents to, and then assess what tools need to be bought off shelf or built from there
in most cases, use existing APIs and then use a harness of the shelf (or build your own) to get it adopted internally
two biggest gaps in this space is product education and forward/applied ai deployment
There's a huge opportunity right now in being the deployment layer for AI into the economy. The amount of work it takes to change out workflows in enterprises tends to be far greater than anyone realizes or would prefer. Clearly this is what the applied layer of AI is going to look like in the form of software and agents, but also it opens up new services firms opportunities.
Legacy systems need to be moved to the cloud, data organization and access needs to be updated, software needs to be connected to agents in new ways, workflows need to be reengineered for agents, HITL needs to be figured out for the process, evals need to be generated and maintained, and the entire system needs to be continually updated as new models get released and new capabilities emerge. And the full list may even be longer.
AI is not the same as just deploying software. Software you generally did the implementation of an existing, well understood category of technology, then stepped back and the customer kept running. With AI agents, you're delivering actual work augmentation to the organization, which has a completely different set of complexities associated with it. You're no longer deploying tools that the company is merely enabled by, you're deploying work output in a process. Completely different implementation and enablement process.
As a result, this is going to open up lots of new kinds of firms and plays for existing firms to diffuse AI into organizations. We're going to see approaches by industry, by size of company, and by problem inside of companies. Traditional SIs will modernize and adapt (some will clearly not adapt as well), and new entrants will also be founded in this period that take advantage of this window. Great time to be an FDE or FDE firm.
Zach Henderson retweeted
Just had lunch with a strategy exec at a 5,000-person company & discussed the biggest AI question they have internally...
Who should own AI within their company?
My answer: we always say you need a boss & a bulldozer.
The boss should be your CEO.
I have never seen a company drive meaning progress if the CEO isn't an involved and supportive champion of the AI agenda at their company. Their job should be to apply top down direction (goals that AI transformation should deliver on), urgency (when your CEO calls something important, you operate with a fire under your ass), and unblocking (CEO has the political capital to move mountains & drive momentum)
The bulldozer is your "Head of AI". Their job is to own the creation & execution of your company's AI strategy, which is defined by 90-day, 1-year, and long-term goals that ladder up to the company's key KPIs.
This person should be full-time. They should not be your CIO, CTO, CHRO, etc. They should literally be your Director/VP/Head of AI in title. They should have full access/permissions to functions & resources across the business with support from the CEO.
The background of this person should depend on where AI will create the most incremental value for the company. If it’s reshaping the product or core technology, their background should prioritize technical depth and product judgment. This could be an ex-product lead, founder, or CTO. If it’s transforming operations, it should prioritize experience leading change across business units. This could be an ex-consultant, COO, Chief of Staff, etc.
But regardless of lean, your bulldozer needs technical fluency, business judgment, and the ability to drive adoption.
Zach Henderson retweeted
Businesses should pay attention to Babak.
Zach Henderson retweeted
Memory vs. Graphs, clearly explained!
memory is great, and the ceiling arrives quietly:
it stores what happened. it does not store what to do about it.
six runs later your file has fifty lines, and the model reloads all of them before it does anything.
Graph engineering fixes this by changing what memory is: not a place things are kept, but an edge that runs backwards.
you need both, and here is the sentence that resolves the whole confusion:
a store keeps what happened. an edge keeps what to do about it.
↳ a store grows with every run, and every line is reloaded before the next one
↳ an edge carries one derived rule, and the rule replaces the run that produced it
Prompts → Context → Harness → Loops → Graphs
the transcript goes away, the constraint stays.
and the constraint is smaller, because "adapters preserve keyword args exactly" is four hundred tokens shorter than the run that proved it.
the same four blocks work on anything you can cut into lanes.
i pointed them at token launches on Robinhood Chain, open source, nothing leaves your terminal
graphrun.pages.dev
the trick is knowing what deserves to survive.
an output is not memory. "ported the utils slice, green on first pass" tells the next run nothing it can act on. the rule you derived from it does.
one thing to know before you scale it.
what you write down is not what comes back.
↳ the root rules file and auto memory are re-injected from disk. they come back intact, every time
↳ path-scoped rules live in message history. they get summarized away and do not return until a matching file is read again
so a rule that must persist cannot be path-scoped. move it to the root and pay the always-loaded cost, or accept that it is advisory in any long session.
and the one that eats whole nights: a memory file that has never had a line deleted is not memory. it is a tax on every run you will ever make, and nobody reads it back.
below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open.
save this, and the repo that runs it is below ↓


























