Leader | Product Manager | Real Estate Investor | Microsoft Data Platform MVP (2017 & 2018)
Joined October 2013
- Tweets 9,534
- Following 574
- Followers 1,340
- Likes 1,244
468 Photos and videos
Jun 7
Will the productivity gain for your median engineer outpace the token cost of getting there?
If the answer is "barely," that is not a reason to retreat from AI. It is a reason to climb.
sqlhammer.com/index.php/2026…
10
Jun 6
McKinsey put the personal AI gain at 20-50% on simple tasks. Under 10% on complex work. Juniors sometimes slower.
That is the ceiling on the vendor harness. The leverage lives one to two altitudes up.
sqlhammer.com/index.php/2026…
16
Jun 2
Earned my Professional Certificate in Designing & Building AI Products and Services from MIT.
The course pushed beyond theory; practical frameworks for turning AI concepts into real products, from ML algorithms to human-machine interfaces to org-level business cases.🚀
46
Jun 2
Multi-agent systems fail in production 41-87% of the time. The model is rarely why. ~79 percent of failures trace to coordination, not capability. Each handoff is a lossy compression step; the agent downstream treats a degraded summary as ground truth.
sqlhammer.com/index.php/2026…
43
May 28
Don't pick your AI model carefully while splitting your agents carelessly. The split is the real decision. Every boundary you draw between agents is a seam where context leaks and work collides. Default to one agent broad enough to hold the whole task.
sqlhammer.com/index.php/2026…
1
66
Apr 11
AI has shifted the bottleneck from writing code to understanding it. The engineer who inherits that codebase six months later pays suffers the comprehension debt that accumulated silently.
Read about the mechanism and the fix.
sqlhammer.com/index.php/2026…
21
Apr 6
AI amplifies your judgment. It does not supply it.
Experts using AI see ~45% gains. Novices with the same tools see 20%. The difference is not the model. It is the human directing it.
The ceiling is set by the quality of judgment behind the prompt.
sqlhammer.com/index.php/2026…
13
Apr 5
Traditional software announces its failures. AI does not. Thrown exceptions tell you where to look. When an AI returns a confident answer that is factually wrong, nothing rings. Measurement is not optional.
New post - Evaluate Outcomes, Not Vibes:
sqlhammer.com/index.php/2026…
1
1
22
Mar 31
Narrower scope produces more focused output. When you define what an AI cannot do, what it does accomplish improves.
Part 7 of First Principles of AI Usage.
sqlhammer.com/index.php/2026…
12
Mar 30
77% of AI-using engineering teams see minimal productivity gains. Carnegie Mellon's Self-Refine research shows iteration alone improves output by 20%, same model, zero upgrades. The model is not the constraint. The process is.
Iterate, Don't Orate: sqlhammer.com/index.php/2026…
20
Mar 30
I asked AI to build a procedural terrain system in one shot. It ran. Unmaintainable. I decomposed the same task into research, proof-of-concepts, and review gates. The delta in quality was substantial.
Monolithic delegation commonly wastes AI capability
sqlhammer.com/index.php/2026…
29
Mar 28
The Amazon Kiro agent deleted a production environment. Not a malfunction; it had permission. The question is never how much to trust AI. The question is whether you've correctly priced the door.
New post on calibrating autonomy to stakes:
sqlhammer.com/index.php/2026…
21
Mar 27
The successful businesses will be the ones that use AI to enhance, not downsize.
tiktok.com/t/ZTkN8oUMB/
8
Mar 27
AI cannot own its output. You can.
Part 3 of First Principles of AI Usage - Why accountability doesn't shrink when you hand work to an agent; it scales with what the agent can reach.
sqlhammer.com/index.php/2026…
13
Mar 26
I love how AI is never offended.
My favorite reply to any of the AI models when some error is encountered.
"That sounds like a you problem."
19
Mar 23
AI does not know what is true. It knows what is statistically probable. Fluent, confident wrong answers are more dangerous than obvious hallucinations. They reach stakeholders before anyone questions them. Your judgment is not optional.
sqlhammer.com/index.php/2026…
1
1
34
Mar 23
You are not just writing prompts. You are constructing the informational environment the AI reasons within. The ceiling on output quality is set before you type a single word.
Context is the product. Part 1 of my First Principles series:
sqlhammer.com/index.php/2026…
16
Mar 14
That was a great Coronet Heist (DnD 5e) one-shot the other night. I am going to find more time to get back into a TTRPG routine.
29
Mar 12
"It must take you longer to make it than it takes for the user to consume it. Anything less is the essence of slop." - Unknown
23
Mar 11
I use Claude Code for my more advanced works with AI. However, a friend recently has been talking a lot about Codex.
What's your take on when to use either?
1
92