EP227: Top 9 Places to Use Jev
A useful cheatsheet of use cases for a decision model like Jev. Also applies to our very own Fernfly.
Bookmarks RSS feed
-
-
Why you shouldn’t wait for inspiration to strike with Harlan Coben
My takeaways:
- Stop waiting for inspiration. Amateurs wait for the muse; pros treat creative work like a job.
- Routines wear out. Coben writes somewhere (a café, a supermarket deli counter) until it stops working, then moves on. Switching pens, notebooks, or screens helps too.
- Get comfortable with boredom. If you take a walk because the blank page scares you, you’re dodging, not creating.
- The magic isn’t the process. It’s seeing words on a page that was blank this morning. Progress beats bliss.
- Character is plot. Don’t describe who someone is; show what happens to them.
- Worst advice: write for the market. By the time you finish, the trend is dead.
- Best advice: if it helps you produce pages, good. If not, bad.
- Good collaboration isn’t 50/50. Aim for equity, not scorekeeping.
- “You bring your own weather to the picnic.”
- Doubt never goes away, even for Stephen King. Let it fuel you, not freeze you.
-
AI Customer Support at Scale: The Travel Industry’s $Billion Bet
In theory, all customer support systems may look the same. This Airbnb vs Booking vs Expedia case study illustrates differing support philosophies that materially affect system design, each with its own tradeoffs.
-
How will AI change operating systems? Part 2: Windows
Cool AI-native OS features by Microsoft, especially around agent isolation/containment and embedded small language models.
-
What is “loop engineering?”
Sometimes it takes a mini-book sized article to learn about a new agentic feature 😅 Nothing revolutionary about loop engineering, useful for low-risk work, one more thing to take humans “out of the loop”.
-
5 More Things World-Class Engineers Do That You Don’t
At Deel, I created a PR to modify a shared database table used across all teams. When I announced my intention to merge it in the developers Slack channel, many people from senior tech leads to CTO jumped in to warm me against it. That felt embarrassing, but I learned more about the tradeoffs and alternatives than I could alone. Collect your battle scars at work. If you aren’t making embarrassing mistakes, you aren’t learning the lessons that will make you elite.
-
ChatGPT vs Gemini vs Claude: How They Differ
Interesting to learn how different companies solve LLM problems like parameter density vs. compute cost, context size vs. output quality, etc., and how they converge on reasoning.
-
The Mistakes You Can’t Take Back
The last one is the most hard hitting - The decision you make by not deciding. The conclusion is worth reiterating - Whichever doors you walk through, walk through them on purpose.
-
Inside Thinking Machines’ Interaction Models
Moving interactivity from harness to model itself is clever, no doubt. What’s more interesting is how they solved constructing responses from partial inputs. And how responses from bigger background model is stitched back in interaction model.
-
How AI Agents Manage Memory and Avoid Forgetfulness
Great deep dive on the deceptively simple concept of llm/agent memory. Biggest takeaway: it’s a game of accurate retrieval rather than storage.