Shankar Bellam

Family ♥ Cloudy skies ♥ Crashing waters ♥ Hopping Music ♥ Founder - SPEQD 💼 Founder - Kāyā Wellness 💼

[Shankar Bellam @shankarbellam]

"Remember who you are" - Mufasa #LionKing

[Shankar Bellam @shankarbellam]

@jun_song Haha! Best gift indeed! That actually gives me an idea 💡

[Shankar Bellam @shankarbellam]

@JesusSoDame PROMPT

[Shankar Bellam @shankarbellam]

@sammgrowth Studio

[Shankar Bellam @shankarbellam]

@jun_song I wonder if the “hook” post can be retweeted in your follow-up? I read that it improves reach for the follow-up posts.

[Shankar Bellam @shankarbellam]

@jun_song

[Shankar Bellam @shankarbellam]

@frankyecom psych

[Shankar Bellam @shankarbellam]

@Mho_23 EDITOR

[Miko @Mho_23]

i just built a 24/7 AI video editor inside Claude Code

this was supposed to stay internal but f*ck it i'm leaking the whole thing..

you give it a reference video and your product info. it comes back with a finished AI UGC ad ready to post.

here's how it works:

the Higgsfield video analyzer MCP breaks down any winning ad you feed it. it pulls the structure, pacing, style, and timing from the reference so you're not guessing what made it work.

then a custom Claude skill takes that breakdown and generates all the video clips using Gemini Omni. which then it renders each scene, stitches them together, and adds captions all in one go.

you're not jumping between tabs or scrubbing timelines or waiting on editors anymore. the whole thing runs inside Claude and outputs a finished 9:16 ad

this is faster than any editor could ever do it and costs almost nothing to run..

i'm sharing the full video breakdown, which includes:

> the Higgsfield MCP setup for video analysis
> the custom Claude skill that handles generation and stitching
> the exact workflow for going from reference to finished ad
> how to batch this out at scale

RT + reply "EDITOR" and i'll send it over (must follow so i can dm)

[Shankar Bellam @shankarbellam]

@jspujji GEO

[Shankar Bellam @shankarbellam]

@AlfieJCarter Agents

[Shankar Bellam @shankarbellam]

@jomatech OMG! This was hilarious and genius! 🙏🏻

[Shankar Bellam @shankarbellam]

@jun_song You are among the few that I have notifications turned on for all tweets. Because every tweet is providing value. Thank you for everything you share with us 🙏🏻

[Shankar Bellam @shankarbellam]

@jun_song Thoughts on Framework Max+ 395 - 128GB?
16-core/32-thread CPU, up to 5.1GHz
40 Graphics Cores, up to 2.9GHz
128GB of non-upgradeable LPDDR5x

https://t.co/KR4VNq6lUg

[Shankar Bellam @shankarbellam]

15 years ago today, I joined X.

A lot of what I know about building, selling, and thinking out loud, I picked up here one thread at a time. So it feels right to start this next chapter in the same place.

I'm building SPEQD. We deploy technology inside medical practices, and we judge it on three things only: patient care, employee satisfaction, and predictable business growth.

What makes it different is simple. I own a practice. Everything gets deployed there first and fine-tuned under real conditions, then offered to clients once it has earned its place.

I'll share the learnings openly from here on out. Call it a small thank-you to a platform that taught me a lot.

Let's build.

#MyXAnniversary

[Shankar Bellam @shankarbellam]

@jun_song Sounds super counterintuitive. You want people to have the maximum power. Especially when rest of the world has it. Or you are crippling your own! I still don’t get it. There’s enough guardrails against “illegal” or “dangerous” usage, so what’s with this paranoia on top of it!🤦🏻‍♂️

[Shankar Bellam @shankarbellam]

@MitcheIl SHOWN

[Shankar Bellam @shankarbellam]

@dabit3 @DevinAI OS for Medical Healthcare

[Shankar Bellam @shankarbellam]

@MitcheIl Please share

[Shankar Bellam @shankarbellam]

@VengeonsP SMART SEO

[Shankar Bellam @shankarbellam]

@jun_song @jun_song 🙏🏻

The real unlock seems to be the harness. Closed teams obsess over agent scaffolding, long-horizon execution, state, and error recovery. Are open labs shipping best-in-class out-of-the-box harnesses so we can immediately build/productionize local agents?