Paolo Valdemarin

Ceo @activateint - we build AI agents to make your business run faster. Investor @activatecapital - helping early stage startups build and launch great products

[Paolo Valdemarin @paolovalde]

A lot of what we write is already being read by LLMs. Let’s make it clearer – for them and for humans.

You know those internal emails that don’t really say much? They’re annoying for us, and they make very poor prompts too.

https://t.co/s6CRWuK2Te

[Paolo Valdemarin @paolovalde]

I picked up my Vision Pro last Friday, as soon as it became available in the UK. These are my first impressions https://t.co/7FvQDPKuWx

[Paolo Valdemarin @paolovalde]

Something to consider while prompting away: LLM can’t think of something and not tell you.

https://t.co/jMjfOt6MHF

[Paolo Valdemarin @paolovalde]

I asked ChatGPT to write a Python script which could run on my mac, access my mailbox, identify new messages since the last run, download each message, use the GPT4 API to read the message and return a value. Based on the value, move the message to a specific folder and mark the message as unread (so I would still see what messages are new in my usual email clients).

It took a few tries but I got it to work nicely. What I find most interesting is that you can tweak the prompt to identify specific types of messages, people, topics, etc.

Eventually I might expand it to respond to spammers ;)

(my knowledge of Python or coding in general extends to copying and pasting code and have some ideas of how it works, all this was AI assisted).

[Paolo Valdemarin @paolovalde]

A friend this morning shared a list of vegetables, noting how hard it is to eat 30 different ones in the same week.

I immediately turned to my AI chatbot to ask to create a list of commonly eaten vegetables, and of course I got a very good one.

At that point I thought that it would be nice to add that list to my next grocery order on @Ocado .

And this is where the magic ended.

Continues here: https://t.co/GT4sgVkQL8

[Paolo Valdemarin @paolovalde]

Coming up with CYA excuses for malfunctions. How human...

[Paolo Valdemarin @paolovalde]

Yet another old fart's post ;-)
https://t.co/TJNz2ZS63P

[Paolo Valdemarin @paolovalde]

@jjenkins_design It makes sense. It’s 24 years since Apple introduced the Aqua UI. Time for a revival. Soon there will be drop shadows, reflections, leather stitches, yay!

[Activate Intelligence @activateint]

Our collaborative project with @GoodMorningIT leverages Agent-to-Agent interactions on the Spritz platform, demonstrating unprecedented efficiency in process automation. Explore our findings and the potential for future applications.

Read more: https://t.co/BJrtJ3b1fQ

[Paolo Valdemarin @paolovalde]

Having just started a company that primarily deals with large language models I’m occasionally thinking about the responsibilities that we have when we introduce a new AI agent in the digital space.

Besides preservation of the human species, a good rule that I think we should give ourselves is “avoid bullshit”, and while this rule must certainly be true for any human activity, I think it’s extremely important when you are dealing with the equivalent of BS thermo-nuclear devices.

I’m still working on my list, this is as far as I got.

Every time one of our AI agents produces an output we should ask ourselves:

does this text improves the life of the intended recipient?
will it be delivered only to the intended recipient (and not to a whole bunch of innocent bystanders)?
is it as efficient as it can be in the way it uses language and delivers its message?
If these minimum parameters are not met, the agent should be destroyed.

As with everything else, AI is not the cause of this, there has been plenty of wasteful content well before a computer could start mimicking the output of a sapiens. And because these LLM models have been trained on a lot of this useless noise, they are extremely good at generating more.

So even before you worry if AI can get the wrong president elected or robots can terminate humanity, just make sure that you are not accidentally leaving the BS tap open when you leave.

https://t.co/ws9I242232

[Paolo Valdemarin @paolovalde]

40 years later: still using a Mac, still excited about what's just around the corner https://t.co/MJ1yNcrnun

[Paolo Valdemarin @paolovalde]

About that moment when every logo is beautiful https://t.co/aolG0YeyZT

[Paolo Valdemarin @paolovalde]

Once upon a time there was the app Moves, which I used for a few years to track everywhere I was going. From Moves I could export data and generate heat maps of my movements. It was great, but then FB bought Moves, and the app died. Strava offers similar maps, which are not as detailed (I don’t track everything with Strava), but still I find all these maps absolutely fascinating. These are the most beaten paths around my two homes.

[Paolo Valdemarin @paolovalde]

"I hope this message finds you well”, where I complain about people that I want to recruit to use AI who actually do use AI. https://t.co/tWfm4W1rau

[Paolo Valdemarin @paolovalde]

For a machine that taught itself to write by watching millions of images the results are incredible, this is not a “and they call it intelligence” post.
I find Dall-E quirky writing fascinating.
Here’s a few motivational posters to get your day started.

[Paolo Valdemarin @paolovalde]

AI and spam: it's gonna get worse before it gets better. https://t.co/28wXXErhNd