Subscribe
Sign in
Home
About me
AI build sprints
Latest
Top
Discussions
How to PM an MCP
Imagine building an AI product where you can’t pick the model, the prompt, or the context.
Jun 3
•
Tal Raviv
36
6
2
May 2026
I don't know what "build an agent" means
I'd rather think in terms of building blocks of LLM applications
May 6
•
Tal Raviv
30
3
1
April 2026
"My PM is using AI to do my job. Is that good?"
—designers, engineers, marketers
Apr 21
•
Tal Raviv
20
1
How to use AI as a team
It's a plumbing issue.
Apr 13
41
14
1
March 2026
Claude is a better PM than me
I'm giving away my Legos.
Mar 17
•
Tal Raviv
95
12
7
We pulled Claude Code into a product design meeting (and accidentally became its subagents)
The best explanation of AI subagents and swarms is being in one.
Mar 12
•
Tal Raviv
18
1
It took me a long time to vibe code a landing page for Familiar
Code was never the bottleneck.
Mar 6
•
Tal Raviv
25
3
February 2026
Introducing Familiar: Never brief your AI again
Install it and your AI gets smarter. Free, open source, and offline.
Feb 16
•
Tal Raviv
48
24
3
My top 54 PM AI guides, in one place, for free
Follow your curiosity, and click on what brings you joy.
Feb 11
•
Tal Raviv
29
1
3
I noticed that Claude Code and Cursor have started managing their own context
Both Claude Code and Cursor context-engineered themselves.
Feb 8
•
Tal Raviv
9
7
The eval that saved our butts: behind the scenes of Lenny's Newsletter
Before going out to 1.3 million readers, our AI had an existential crisis.
Feb 5
•
Tal Raviv
12
A flaw in our Lenny's Newsletter guest post
It's an open-source, locally-running AI product—and no observability.
Feb 3
•
Tal Raviv
9
2
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
