@ericries Great episode, Eric. May I suggest that you interview Pieter Viljoen from Allianz who is overseeing probably the largest agentic transformation globally? One of the “brave” putting AI into delivering real value.
Anyone who said agents “execute every single task with perfect consistency regardless of how many times they've done it” clearly don’t work with agentic in Enterprise.
Don’t get me wrong; I love my Openclaw assistant, but it’s a looong way from enterprise consistency.
At least one Padres fan in the UK… perhaps not many more than that! (Who else leaves SD for England?!)
One of my first memories is of the ‘84 series. And completely remember watching the Cubs after school with WGN on cable #basescoveredlive
@mattfrei@virginmedia Do you use iOS/macOS? Is Private Relay dropping consistently? The parental blocking on the virgin router doesn't play well with Mac, it seems (after months of debugging myself and, finally, a breakthrough via GPT-o3 in reviewing logs).
Turn off parental blocking. Voilà.
Fantastic interview questions.
Also, I love the idea that coding skills are table stakes (with GenAI levelling that table).
And, for the record: I can explain my home WiFi in some epic detail.
Your technical interview process is wrong.
After building multiple engineering teams at Microsoft, VMware & Instacart,
I've discovered what really predicts performance.
Here are the 3 traits of the best engineers:
@michelifelse Everything I’m doing is about automating back office operations. Lots of use cases; many/most centred around manipulating information in and out of Office documents whist making semi-rule-based decisions.
How far are we from having competent AI co-workers that can perform tasks as varied as software development, project management, administration, and data science?
In our new paper, we introduce TheAgentCompany, a benchmark for AI agents on consequential real-world tasks.
A weird and interesting thing about the new ChatGPT Canvas mode is that it can run Python in an entirely different way from the existing Code Interpreter - using Pyodide (Python in WASM) - which means it can make network requests from Python now!
I've yet to see a good thread summarising new types of UI that generative AI has necessitated, so here's an attempt to start one (in the hopes you all can contribute!)
Not necessarily endorsing any of these, I just see them all as relevant in our future
Let’s see if we can crowdsource a robust definition of “agent” (with respect to AI and LLMs) that fits in a <=280 character tweet
Reply to this with your best attempt, then scroll through the replies and fave the ones that makes sense to you
Sat at dinner in Cornwall with a group of friends, trying to explain the legend @timtenyen circa 2010 in Camden. Playing Sea Anemone doesn’t do it justice, but we’ll start with that before moving to the Bear and the Fox.