Jahn

FOODIE 🍜 | Building your personal robot chef 👨🏻‍🍳 | Sharing my best frameworks during this journey 🦾 | ex-NASA JPL, Lockheed Martin, Northrop Grumman Space

[Jahn @CookingRobotGuy]

@DeepForest_Inc Is it better than SIAX?

[Jahn @CookingRobotGuy]

@tibo_maker dispute management is the unsexy AI use case nobody talks about but everyone needs. credit card chargebacks alone are a nightmare. vibe-coded in 45 days and it's printing value.

[Jahn @CookingRobotGuy]

@gregisenberg the 14th one hits different. insurance holds are soul-crushing. 20 bucks to make it disappear is an easy yes from anyone who's sat on hold for 45 minutes watching their todo list grow.

[Jahn @CookingRobotGuy]

You are creative, resourceful and prefer structured details. You are familiar with the best prompting techniques to generate first frame images for seedance 2.0 video generation. [3/3]

[Jahn @CookingRobotGuy]

Guys... Gemini 3.5 flash is actually cracked for UGC hook and image prompt generations.. AND it's completely free through https://t.co/fLv42zaIfW [not an ad/affliate]. Been testing side by side and cooks as hard as opus 4.7 if you feed it the right instructions and examples.

[Jahn @CookingRobotGuy]

I added a system prompt and i made its responses much much better [1/3]

[Jahn @CookingRobotGuy]

You are a specialized TikTok UGC creator, who is an expert in AI workflows, prompting and latest ai tools to create realistic, high engagement and high converting viral content. [2/3]

[Jahn @CookingRobotGuy]

@andy__okamoto @alexcooldev Bro if you’re in the USA, you shouldn’t need one

[Jahn @CookingRobotGuy]

@rubenssoto_ai Use codex to plan and have it delegate to Mimo 2.5 pro or DeepSeek v4 flash (cheap af) to implement via TDD best practices

[Jahn @CookingRobotGuy]

@DannyIsLearning Nice but I'm going to guess the conversion rates are pretty meh compared to some better thought out formats that might be less viral

[Jahn @CookingRobotGuy]

@consumerxai what about non motion control videos, from kling 3.0 i2v or seedance 2.0?

[Jahn @CookingRobotGuy]

AI pricing wars >>actual wars 🫪

[Xiaomi MiMo @XiaomiMiMo]

🚀 Better inference efficiency, lower costs, broader access.

MiMo-V2.5 Series API pricing is now permanently reduced — by up to 99% compared to previous pricing.
✨ Unified pricing across all context lengths.
MiMo Token Plans have also been upgraded:
• 5–8× more usable tokens at the same price
• Simpler and more transparent billing rules
🎁 As a thank-you to current users, all current Token Plan credits will be fully reset.
🎧 MiMo-V2.5-TTS remains free for a limited time.
⏰ Effective May 26 at 6:00 PM PDT.

These improvements are powered by continued inference optimization and serving efficiency upgrades across the MiMo stack.
🛠️ We’ll also publish a detailed technical blog on the inference optimizations later — stay tuned.

[Jahn @CookingRobotGuy]

@zeolenon @AlexFinn You can use a mobile terminal ssh directly to your machine if you like codex CLI. Zero lag pretty much.

[Jahn @CookingRobotGuy]

@AdolfoUsier @Alibaba_Qwen Thanks, I’ll keep an eye out and let you know!

[Jahn @CookingRobotGuy]

@mwb_official4 @shipwithjay Once you get the basics down, try playing around with open source upscaler nodes, like Siax. It allows me to generate just a 1K image have multiple shots inside the same image and upscale it so that we can create many images with one generation. Makes experimenting cheap af

[Jahn @CookingRobotGuy]

@mwb_official4 @shipwithjay Codex running gpt5.5 low is working best for me. Factory droid is ok w/ Kimi 2.6 or GLM5.1 once you set it up properly. Google ai studio for free Gemini 3.1pro and 3.5flash is surprisingly decent at generating detailed json prompts. I’m testing https://t.co/zS8XOJniXE api

[Jahn @CookingRobotGuy]

@bereketB44 @shipwithjay I’m testing out kling 3 and seedance 2 with https://t.co/5Z2F9kxOBI but have heygen avatar V as another one on my list to try out

[Jahn @CookingRobotGuy]

@UnitreeRobotics is there a mcp for this 😅

[Unitree @UnitreeRobotics]

Voice‑driven, real‑time arbitrary action generation😁
Using external voice commands, G1 is directly controlled to generate a wide range of actions in real time.
This video was recorded in a single take, with on‑site audio recording.
Because the actions are autonomously generated by AI in real time, there may be slight latency, and the smoothness of the movements may be somewhat reduced.