i don’t use the ai overview (except to laugh at them); the web links are mostly fine
LLMS
-
Workshop: Build Reliable GenAI Apps with Evals and Observability
By
–
Workshop hosted by @PacktPublishing @PacktDataML — "Build Reliable GenAI Applications with AI Evals, Observability, and Testing" 𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗵𝗲𝗿𝗲 with my discount code 'KIRK40' (𝟰𝟬% 𝗢𝗙𝗙 already applied): https://
eventbrite.co.uk/e/build-reliab
le-genai-applications-with-ai-evals-observability-testing-tickets-1987301251540?aff=KirkB&discount=KIRK40
… 𝗟𝗲𝗮𝗿𝗻:
How production AI -
Loop engineering is real; private evals become your moat
By
–
Loop engineering is REAL. It’s not about picking the best model. It’s the loop you build on top that compounds. Private Evals are about to be HUGE. Score models on your outcomes and your traces, not a public leaderboard. Your EVALS are your moat.
-
The model is the ecosystem; capability is fundamental
By
–
the model is the ecosystem – always has been, always will be! everything else just helps the model get to an outcome faster/ cheaper tho nothing helps if the model itself is incapable
-
Every model can be jailbroken, despite appearing easy
By
–
It's a lot harder than it looks. Every model can be jail broken.
-
Deep research better with ChatGPT according to tunguz
By
–
Deep research is still better with ChatGPT IMHO.
-
No financial incentives for open weights frontier models; huge capability gap.
By
–
The financial incentives for training frontier open weights models don’t exist. Small models, sure, but there is a huge capability gap.
-
The excessive cost of tool calls in agentic coding
By
–
Biggest problem in agentic coding? The tool call tax! (No, it's not an EU regulation.) Did you know… – It costs you *10x* more $$$ to generate 100k tokens that include T=30 tool calls vs. T=1 tool call at the end of your
-
No Fable 5 jailbreak against Opus 4.x or GPT 5.x, fortunately
By
–
I'm just glad that no one in the US government thought to try this Fable 5 "jailbreak" against Opus 4.x or GPT 5.x, otherwise I would have absolutely nothing useful to do this weekend.
-
AI generates a personalized hook for each company in 2 minutes
By
–
AI reads each company. Understands what it does. And writes a hook that really references their activity. Not "Hello {{First Name}}, I noticed you work at {{Company}}." Real personalization. At scale. In 2 minutes. What it…