Benjamin Starling

[Thorsten Ball @thorstenball]

Who's in NYC next week?

Let's meet!

https://t.co/TpNw8JE0G4

[Benjamin Starling @bstarling_]

@thorstenball I’m in!

[Benjamin Starling @bstarling_]

@lindsmccallum Alfred!

[Benjamin Starling @bstarling_]

@jxnlco I need my dot temporarily so it can manage a custom solution that will bit rot instead of improve over time.

[Benjamin Starling @bstarling_]

@IanLandsman Maybe for b2c but @AmpCode puck is an example of an "agent" that is very useful for coordination/handoff. I think dots or other similar agents when fully implemented can be great as the point of contact when you're on mobile or managing a ton of projects at once.

[Benjamin Starling @bstarling_]

@thsottiaux More than one computer attached at a time. Hard to keep all the conversations straight on mobile (better threads? separate channels?). iPhone shortcut to dictate directly to your dot.

[Joel Grus 🤠 @joelgrus]

asking people at the gym how excited they are about openai dots

modal answer "what the hell are you talking about"

[Benjamin Starling @bstarling_]

Testing out the new openai Dots. It was creating thousands of temporary projects files in my Documents folder which was furiously syncing to iCloud. 5gb in total before I noticed what was happening 🙃

[Beyang @beyang]

Well, not everyone

[charlie blank @charlieblank_]

i am literally always asking muse / instinct to book me a private jet and it is so frustrating when it makes me log into gmail to complete - this is sick, lfg!

[Benjamin Starling @bstarling_]

@sibaiisaac Do you honestly believe that someone/a team of someones looking out for the community there?

I agree "leaving it alone" is better than ruining it but that is a pretty low bar? Clearly the site is slowly degrading with half working features

[Benjamin Starling @bstarling_]

Agree, Goodreads could be an amazing product. I wish it was well taken care of. I still use it regularly but there are so many broken/half working sites without even starting on all the cool stuff that could be done.

[Sam Parr @thesamparr]

I wish I owned Goodreads.

It has SO MUCH potential. Yet, sucks in so many basic ways.

Has the hardest part: tons of great reviews and a good community. Everything else...horrible.

Search, list making, claude integration, so many more.

[Benjamin Starling @bstarling_]

Nice, gonna have my agents read this before deciding how hard to think 🤣

[Thariq @trq212]

What is effort really? When do you change it it and why not just use max effort for everything?

I dove deep into this problem, looking into evals and doing my own tests and I was quite surprised by the results. https://t.co/KO2D51j34H

[Benjamin Starling @bstarling_]

@PlusEVAnalytics With something like amp code orbs or soloterm it's easy to tell agents to argue or verify each other with a simple prompt.

"Spawn a solo (orb) sub agent to do X work, verify this is done with a screenshot"

https://t.co/XwdBbpkg8M
https://t.co/QmYrq9B3WY

[Benjamin Starling @bstarling_]

How do I teach my twitter algo this is what I want?

[Aaron Francis @aarondfrancis]

Things are changing, never doom.
Jobs will look different, never doom.
Software is being eaten, never doom.

Increase your ambition, care deeply about your (changing) craft, build things you're proud of.

Never doom!

[Benjamin Starling @bstarling_]

@aarondfrancis What do you say if they are in trouble though

[Steve Yegge @Steve_Yegge]

Can we all just pause for a sec and reflect on how fucking important Tailscale is, and how nobody ever talks about them? I don't think I've ever seen a more useful service with a more invisible corporate footprint.

[Benjamin Starling @bstarling_]

I have the same issue anytime I try to switch to dictation. I keep trying every few weeks though.

[almonk @almonk]

idk about anyone else, but I find it really hard to dictate what I'm thinking. typing unlocks something in my brain that speaking just doesn't.

[Benjamin Starling @bstarling_]

@netcapgirl Circle back when complete

[Benjamin Starling @bstarling_]

@iannuttall Makes me wish I had my own set of benchmarks so I could get a better read on stuff like this without having to search to see if people here agree.