Tom Delangle

Exploring AI agents in public. Honest beginner, sharing what works and what doesn't.

[Tom Delangle @DelangleTom]

Bon bah all in sur Higgsfield à ce prix là hein 🥹

[Higgsfield AI 🧩 @higgsfield]

Introducing Higgsfield API.

50+ frontier models in one API, at lower prices than a subscription.

> Get up to 50% OFF discount on your 3 favorite models
> Lock in your max-discount within 7 days
> Pay per use with no commitment

Build your own Higgsfield with the best prices in GenAI industry.

Available at https://t.co/Ow8CcuTQ1s

[Tom Delangle @DelangleTom]

@guidorosso @kueiniu @rive_app Amazing*

[Tom Delangle @DelangleTom]

@guidorosso @kueiniu @rive_app Amazon news, can’t wait to try it after work. By the way, love you product!

[Tom Delangle @DelangleTom]

@kueiniu @rive_app A whole new world to explore then 🤭

[Tom Delangle @DelangleTom]

@tonysimons_ Is the virtual iPhone able to install any app from App Store. Curious about this repo, thanks for sharing.

[Tom Delangle @DelangleTom]

On y croit vraiment à GPT6 qui te sort des trucs de dingo sur Blender, ou bien bullshit ? Intéressé par des retours à chaud !

[Tom Delangle @DelangleTom]

@familystyle46 @buildkevins J’utilise ça pour automatiser une partie du taff / ça fait bien le job. J’ai pas testé si ça te permet de remplir les formulaires directement via ton Claude code/cursor : https://t.co/pWRHiQ90bV

[Tom Delangle @DelangleTom]

@buildkevins Comment tu fais pour que TikTok localise ton compte aux US ?

[Tom Delangle @DelangleTom]

@adrianabelarde_ Too much

[Tom Delangle @DelangleTom]

Bon, les gars, je viens tester l'App Store Connect CLI de RORK et j'avoue, c'est juste une dinguerie. En début d'année, j'ai mis au moins 2/3 journées pour tout capter, mais maintenant Claude Code fait tout, c'est incroyable. Tflight + push review en 1h...
https://t.co/pWRHiQ90bV

[Tom Delangle @DelangleTom]

@filippocarnvl « Couple of prompts »

[Tom Delangle @DelangleTom]

@Eloise1041955 Tu as une app (How we feel) qui se rémunère via des dons. Peut-être une idée ? Si tu expliques bien les besoins, la valeur du produit. Ça peut passer ?

[Tom Delangle @DelangleTom]

Spent the morning using autoresearch on itself.

My laptop has a 4GB RTX 3050 Ti, so the default setup just OOMs. I made it run a small sweep, test configs one by one, and pick the best setup for my actual hardware.

very meta. also very useful.

[Andrej Karpathy @karpathy]

oh yeah i should have linked autoresearch probably
https://t.co/YCvOwwjOzF
(you don't "use it" directly, it's just a recipe/idea - give it to your agent and apply to what you care about.)

and the tweet about it that went mini-viral over the weekend with more context
https://t.co/q5eWsvx5p2

[Tom Delangle @DelangleTom]

Question for Hermes people:

Are memory providers supposed to replace Obsidian, or are they different layers?

my guess:
memory = tiny facts the agent should recall
Obsidian = artifacts humans can read/edit/link

anyone routing different memory types to both?

[Tom Delangle @DelangleTom]

@TheBenValentin Thanks Ben, that’s exactly what clicked for me!

[Tom Delangle @DelangleTom]

Feels like the interesting part isn’t just bigger models anymore.

It’s personal AI that can actually see, listen, use tools, remember context, and act inside your workflows.

I’m still early with Hermes Agent, but this is exactly the direction that makes me want to learn agents!

[Thinking Machines @thinkymachines]

People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way.

We share our approach, early results, and a quick look at our model in action.

https://t.co/AFJZ5kH7Ku

[Tom Delangle @DelangleTom]

Hi @NousResearch, is it intended that the WhatsApp mode can't be changed via hermes setup / hermes whatsapp? Feels like a shame to have to edit the config files by hand. Would be great to add that step directly into the commands.

[Tom Delangle @DelangleTom]

@seraleev Any cheaper workaround? I find it expensive when you want to iterate on a video…