Korra - Got Data? retweeted
cheapest way to learn reinforcement learning:
- David Silver's RL lectures (free)
- implement multi-armed bandits
- code Q-learning from scratch
- implement DQN on CartPole
- learn policy gradients and PPO
- train an agent in Gymnasium
- read the GRPO paper and understand how it relates to LLM post-training
you can learn most RL fundamentals on a laptop.
resources in replies.
Korra - Got Data? retweeted
Replying to @AnthropicAI
My 8-year-old son Adam has been fighting Shiga toxin–associated hemolytic uremic syndrome (STEC-HUS) in the ICU for 52 days. What started as an E. coli infection became a life-threatening illness affecting his kidneys, brain, and multiple other organs.
Watching my child fight for his life has changed me forever.
I’m a software engineer and technology company founder, and I’m willing to dedicate my time, resources, and technical expertise completely free of charge to help researchers use AI to develop better treatments for this devastating disease.
@AnthropicAI @arnavmehta3 — if your drug discovery initiative could help advance research into STEC-HUS, I would love to contribute in any way possible.
I don’t want another child or family to experience what Adam is going through.
If we can turn his suffering into progress that saves even one child’s life, it would mean everything to me.
Korra - Got Data? retweeted
Replying to @gauravsunil_
Yeah the Neo labs are pros at creating hype by offering 50k referral fees / nobody asks how many people got paid relative to the amount of deals that were submitted….
That number would scare a lot of brokers away.
Korra - Got Data? retweeted
Holy crap, who the hell did this?!!
Someone open-sourced a UGC video studio and lets you run it on your own computer.
GitHub's already at 4,047 stars, repo's called OpenShorts.
You just paste your product link, and it writes the script, creates an AI actor, gives them a voice, and syncs their lips. B-roll, subtitles, hook overlays - it's got all that covered too.
And it's not just ads. Feed it a long video and it'll find the good bits, follow the speaker's face, and turn them into vertical clips for TikTok, Reels, and Shorts. Thumbnails, titles, descriptions - they threw those in too.
No camera, no filming yourself, no watermark on the self-hosted version
Link's right here: github.com/mutonby/openshort…
The catch? AI actors and voices still use paid APIs.
The studio runs on your machine; those models don't.
But hey, the studio itself is free, and you can go from a product link to a talking spokesperson without picking up a camera. Who needs a damn film crew!
over the past year i've helped dozens of people grow their personal brand on X & YT...
i've built a 7fig/year business off the back of it without charging anyone a SINGLE dollar
if you wanna work with me, i'm making a free group chat for anyone who thinks they can benefit from my help
requirements:
> must be building on whop, or planning to
> must be high agency
> must have the time to make content
what's in it for you
> free content strategy assistance
> a free launchpad for your personal brand
> monthly rewards for the best content
i wanna build a network of 100+ people building & documenting their journey publicly in a way others can benefit from
if you're interested in joining, RT this tweet & reply "i'm in" & i'll DM you
Korra - Got Data? retweeted
Stated preference and behavior does NOT align.
Just realized I love when an Uber doesn’t talk
But every time I’ve fat tipped one, they were talking & we made a connection.
My stated preference is not what I’ve rewarded
Korra - Got Data? retweeted
Neil Movva (@neilmovva) started his career at Nvidia, working on GPUs and kernels, and has an unusually deep understanding of inference, from software to chips to power.
We spend a lot of time on each of those layers, how they connect, and where the important tradeoffs are.
What makes this conversation special is how detailed it is (like a 401-level class), yet Neil makes it remarkably clear and easy to follow.
Today he runs Sail Research, a company building infrastructure for agents to make tokens as cheap as possible.
We discuss:
- Latency versus throughput
- Why there are no bad chips, only bad pricing
- The end of kernel engineering
- Buying chips and power no one else wants
- New chip architectures
- Nvidia lore + his contrarian view of the company
- Open source and the frontier labs
I learned a ton. Enjoy!
TIMESTAMPS
0:00 Intro
0:38 Building a “Token Factory”
4:21 The Future of Background Agents
13:09 Nvidia and the GPU Stack
23:27 Chips, Memory, and Transformers
36:14 The Future of AI Training Data
44:32 Chip Scarcity and Compute Arbitrage
52:44 Reinventing the AI Data Center
59:01 Power and the “Scavenger Strategy”
1:10:10 Open vs. Closed AI
Korra - Got Data? retweeted
Being cringe is the tax you pay when you start to prove to the world that you care more about what you want than how you look.
Korra - Got Data? retweeted
i built a way to generate full ai vlogs for youtube using seedance 2.5, without ever picking up a camera
this was supposed to stay internal but f*ck it, i'm leaking the whole thing
you give it a script or even just a topic and it comes back with a finished vlog, voiceover, b-roll, captions, ready to upload
here's how it works: a custom claude skill breaks your script into scenes and writes timestamped storyboard prompts for each one, matched to pacing so nothing feels random. those prompts get fed into seedance 2.5, using its 30 second continuous generation and reference asset support to keep the visuals consistent across scenes instead of looking like ten unrelated clips stitched together. then it pulls in an ai voiceover that matches the tone of the script, stitches every scene together, and burns in captions automatically
you're not filming, not voicing, not editing, not exporting from three different apps
the whole thing runs start to finish and outputs a finished 16:9 vlog ready to post, and it costs a fraction of what an editor or camera setup would run you
i'm sharing the full breakdown, which includes:
the claude skill that handles scene breakdown and storyboard prompts the seedance 2.5 setup that keeps visuals consistent scene to scene the voiceover setup that keeps it from sounding robotic the exact workflow from script to finished vlog
RT + reply "PROMPTS" and i'll send it over (must follow so i can dm)
Korra - Got Data? retweeted
Today we’ve raised $52M Seed and we are announcing the public launch of S2.1 Pro.
>It can clone a voice from 5 seconds of audio
>2x faster than Cartesia & 1/6th the cost of Eleven Labs
>most expressive model with word level control over emotion, intonation, pacing etc
We support frontier AI companies including HeyGen, LiveKit, Retell, Sanas, and OpenArt all run our model in production.
If you're a business and we can't cut your voice AI costs by 50%, we'll give you 1 year of Fish Audio for free.
Book a demo: s.fish.audio/tmapke
To celebrate our first birthday, we'll give you 1 month of S2.1 Pro for free. Like, retweet, and comment “Fish” to get it.










