Jepey

[Jepey @capin_paul]

@Riel_Chou @callmesenseieng That combo is amazing!!

[Call Me Sensei @callmesenseieng]

your Sensei. your chat. your screen.

Sensei stickers + wallpapers are in Call Me Sensei. send them in Messages, set them behind your chats, and play Read the Mood for a Toma sticker ✦

https://t.co/p9vjzcZQEO

[Jack Vinijtrongjit @juminoz]

Introducing Crushfall, a new Roguelike puzzle mini-game for @callmesenseieng. The game was inspired by one of my childhood favorite Street Fighter Puzzle 2 Turbo.

Crushfall will be available within the Call Me Sensei app on Monday October 5th.

You can access it once the mini-game feature is unlocked so if you have already progressed beyond that, you will have an access to it right away.

15+ mini-games are also on their way and we will be available soon as well.

[Call Me Sensei @callmesenseieng]

Sensei: “we need to talk”
Jude: what if we actually don’t

[Call Me Sensei @callmesenseieng]

relationship advice nobody asked for but somebody needed 😭
and if he STILL doesn't get it… maybe it's time to see who you're actually compatible with 💀 take the quiz and meet your match — link in bio 💜

#CallMeSensei #DatingHumor

[Call Me Sensei @callmesenseieng]

we’re not homeless we’re ✨fully remote✨

turns out all a startup really needs is 4 laptops, free cafe wifi and somewhere to sit 🥰

#startuplife #startup #remotework #coworkers #callmesensei

[Jack Vinijtrongjit @juminoz]

Along with the launch of hosted version of ToonLab tomorrow (Monday Japan time), we will also launch a new anime production platform. Let me know if you are interested in giving it a spin. We have had many people asking us to productize our harness and I finally got around to do it. Some of the more ambitious features such as book/manga-to-movie and series will also be available, but very experimental. Technically, you can one shot a 90-minute movie if you are brave enough (and okay spending some money of course). The QC process is in place to help with quality control, but it also means additional generations.

Anyway, more details tomorrow.

And if the @github god decides to give me back my account, I will also release the open source version of ToonLab. If anyone knows someone from their team, please connect me. I have tons of projects I can't access because I recently got a new laptop and didn't sync all of them. It started to become very challenging to do things.

[Call Me Sensei @callmesenseieng]

Call Me Sensei is officially in OPEN BETA! 💜 Call Me Sensei is an immersive language learning dating sim where you can build relationships with your Senseis while practicing language through natural conversations and experiences that become part of your daily life.💬 #OpenBeta

[Jack Vinijtrongjit @juminoz]

Just published a song for the first time. The song is for @callmesenseieng and it's called「声の先へ」ー Koe no Saki e ー Beyond the Voice.

The process was way easier and cheaper than I thought using @DistroKid. If all goes well, the song will be available on all major platforms sometime within the next 72 hours.

[Call Me Sensei @callmesenseieng]

the quiet ones are always the most dangerous.

would you fall for ren? 👀

#callmesensei #ren #anime #animeboy #otomegame

[Jack Vinijtrongjit @juminoz]

Throughout the journey of developing @callmesenseieng, we have developed tons of tools (50+) and people who have seen our admin console has been asking us to expose some of our tools. So here's the first one. It's called "This Is The Shit".

This is an internal tool that we typically use to evaluate marketing videos. It's deterministic so if you upload the same one, you get the exact same scores and evaluations (yes, LLM can be 100% deterministic if you know what you are doing).

Anyway, we thought we put a fun spin into it and allow people to submit their videos for verdict. Once evaluated, they stay up for everyone to see. If the video is good, it goes on "The Shit" list. If bad, it goes on "Shit" list.

The site will also provide useful information in a humorous way and will continue to provide the most up-to-date trends, especially how AI is being used and how it's impacting marketing in general.

Anyway, to celebrate the launch, the first 100 people who sign up, can use this coupon code at checkout for 1-month subscription. By default, you can also test it out with 2 videos for free.

Code: IAMTHESHIT

https://t.co/KFgheJFQJf
https://t.co/z2KfBA5OM3

[Jack Vinijtrongjit @juminoz]

Progress after 1 hour and 46 minutes. It now supports WebGPU and the surfer is actually surfing. The wave is also looking pretty good.

I just found out that the actual wave implementation file was missing from the source code so a lot of work has been done via approximation.

So far this has take 18% of 5-hour budget and 3% of weekly budget.

Environment: GPT 5.6 Sol XHigh Fast with 544,000 context window inside Codex environment

#surforturf

[Jack Vinijtrongjit @juminoz]

Major improvements on lighting, shadow, and general environment shaders for outdoor and indoor scenes. The goal is to support both objects with texture and without texture. In this example, the wooden bench is a @quixeltools MegaScan asset and I have set it up so it will auto detect MegaScan assets and apply base processing to make sure they fit into stylized environments.

I have always seen a lot of impressive examples using @threejs, but I would say this is way beyond my expectations on how far I could push.

I will now focus on bug fixes and additional procedural generations for foliage. Once that's done, I will package it and push it out to NPM and will likely put up a website to showcase various art styles created by the community. Will share the plan on how the open source project will be structured soon.

[Jack Vinijtrongjit @juminoz]

Text-to-animation definitely still has a long way to go, but you can now iteratively prompt with a model like @AnthropicAI Fable 5 to get the animation you want. The experiment was only for the freestyle swimming part.

Summary:
- Task: Freestyle swimming with head turning to get some air every 4th stroke (ended up with every 2nd stroke, which is fine also)
- 3% of weekly Fable allowance was used
- Estimated cost: $1.20
- 7 iterations because Fable's vision is so terrible (please fix it @AnthropicAI). I had to keep taking screenshots myself and then describe every painstaking details of what's wrong with it, but it did get there eventually.

And yes, all the shaders, water mechanics, tree, grass, etc. you see here will be open sourced soon (1-2 weeks). This is being developed as a mini-games engine based on @threejs for @callmesenseieng (Open Beta available soon).

[Call Me Sensei @callmesenseieng]

POV: You're the only marketer in a startup. When you try to join the dev team discussion. #MarketingStruggles #StartupLife #POV #Relatable #FunnyMeme

[Chetaslua @chetaslua]

Seedance 2.5 is mythos moment of Video Models

> more than 30 seconds
> 4k resolution
> prompt adherence is on another level that we never seen before
> our reports was right , this is why follow us and join our server
> July will be do much fun

[Call Me Sensei @callmesenseieng]

The matcha machine is down for a week.

We’re asking everyone to respect the team’s privacy during this difficult time. 🍵🕊️

#OfficeHumor #StartupLife #MatchaAddict #CallMeSensei #StartupCulture

[Kōda @aimikoda]

Anime character styles are incredibly diverse and researching them one by one can get exhausting fast. I'm using a simple GPT Image 2 prompt that turns this process into a visual catalog by generating infographic-style posters with different anime character archetypes, visual languages and era-specific styles all in a single image.

It is a quick way to explore, compare and study character design directions without jumping between dozens of references. Great for inspiration, style research and building visual libraries.

You can tweak a few parts of the prompt to explore other directions as well, such as stylized 3D animation looks, cinematic visual languages or even film-inspired design catalogs.

Example Prompt: Create an English infographic catalog poster showing 1990s anime character styles in a clean labeled grid layout, each panel featuring a different archetype and visual language. Use 16:9 aspect ratio.