HeyJaredJorge Alcantara
Building PM tools after a decade of implementing AI globally (Microsoft, Samsung, 200+ others). Sharing honest insights about product development in the AI era. SF & Madrid based. Always exploring how to help teams build better things, faster!
- Indexed issues, last 90 days
- 3
- Latest publication
- Sep 9, 2026
- Audience
- Checking…
- Earliest in this view
- Aug 25, 2026
Latest issues
Choose Context Before Capability (opens the original)
Read excerpt
An agent can read “add a CSV export” and create a ticket. The harder part is finding which fields finance agreed to, and why.I build Zentrik, which appears below. This comparison draws on the linked vendor documentation.<source type="image/w
A life that lucks. (opens the original)
Read excerpt
At least five of them are attempts to explain a Claude Code cache & harness mechanics. 3 are “Untitled” and have a wild mix of ideas. One is literally a dump of limits and no-nos called “Jorge voice control”.No sensible publication needs this many tests. Mine apparently did.Thanks for reading! Subscribe for free to receive new posts and support my work.<input class="email-input" name="email" tabindex="-1" type="e
Changing the reasoning effort is costing you money (opens the original)
Read excerpt
If, like me, you’ve been trying to get the most out of your Claude Code and Codex subscriptions, you may have been switching models and reasoning efforts constantly to min-max token vs. efficiency.And, like me, you may be leaving tokens on the table.A few days ago, due to the talk in reddit about lower credits on Codex, I went on a rabbit hole, trying to understand exactly how to optimize cache usage in your interactions with the main harnesses.This is not always behaving exactly as you expect b
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.