Worth Calculator scores your job based on salary, hours, commute, environment, and personal factors — not just your paycheck. https://x.com/GithubProjects/status/2078442695278871038

Confession: I’ve been sleeping with my laptop for the last year

Another classic example of how Indian workplaces operate!!!

Yesterday, I posted about how an Indian male employee asked his US manager to be OOO for 3 hours daily to take care of his baby.

Here, a female employee from Zomato asked her TL for a WFH because of her health issues.

But instead she was harassed mentally and sexually. And when she tried to complain, Zomato terminated her and not even issuing her an experience letter. Even Zomato HR has become deaf in this case.

If a known startup Zomato's work culture is like this, what will you expect from other companies?

This is who runs this account https://x.com/nikitabier/status/2078610519926423710

llevo un rato dándole vueltas a esto y no me lo creo

Seedance 2.0 on Higgsfield AI
Prompts y assets open-source completos abajo:

o sea el prompt que usaron ELLOS para ESTE video, te lo ponen ahí para que lo copies, así de fácil

antes tocabas adivinar cómo lo habían hecho
ahora te lo dan todo hecho

@Image1
is the lead performer — preserve her exact identity in every shot: cornrow braids, septum ring, statement earrings, sculptural glossy white wrapped top, dark indigo denim.
@Audio1
is the music track — entire edit cut to it; no invented music, no voiceover, no lip-sync.

Music-video route: dance/performance, rooftop golden-hour editorial. Track: industrial hip-hop, 95 BPM half-time, distorted 808s, luxury-noir. Director thesis: a skyline coronation — the lead and her four dancers own an empty rooftop at last light, silhouettes against the city.

Visual world: flat concrete rooftop of a brutalist tower at golden hour, low sun raking long shadows across the roof, city skyline hazy in the background, wind moving fabric. Background dancers: 4 dancers in tonal rust and camel utilitarian looks, sculptural streetwear. Palette: golden amber, rust, concrete, bone white top cutting through, deep indigo. Natural light only, no neon, no particles.

Cut map synced to
@Audio1
:
0.0–2.0s THE BREATH — static ultra-wide from across the rooftop: five figures spread far apart in a pentagon, all facing the sun, long shadows pointing at camera. Wind and stillness.
2.0s THE DROP — 808 kick: all five pivot to camera in unison on the hit; hard cut to low-angle medium of the lead, sun flaring behind her braids.
2.0–4.0s she struts along the roof edge parapet line (safe, well inside the roof), the four dancers converging on her diagonally from the corners in lockstep, shadows sweeping like clock hands; camera dollying backward.
4.0s snare — whip-pan: full freeze in a tight cluster, she hits a profile vogue arm frame, dancers frozen in four different angular silhouettes against the skyline.
4.0–8.0s golden-hour choreography in medium-wide: her sharp head isolations and New Way vogue arm lines land on the snares; the four dancers frame her in rotating formations — box, diagonal, split pairs — changing formation exactly on each snare hit; torso rolls ripple through the group on the 808 slides. Camera orbits with the sun cross-lighting faces.
8.0s hi-hat triple roll — 3 rapid macro inserts: her earring flaring in backlight, four hands hitting the same vogue frame stacked in depth, boots pivoting on rooftop gravel.
8.5–12.0s escalation: accelerating orbital around the cluster as the dancers whip around her in a fast carousel, her solo arm work speeding up at the axis, long shadows strobing across the lens; speed-ramp slow-motion stretch — hair and fabric floating in wind — then snap back on the kick.
12.0s bass slide — the whole group drops level together into deep sliding lunges across the concrete, camera dutch and low, the sun skimming the roof surface.
13.5–15.0s FINAL BUTTON — hard cut to the opening ultra-wide: pentagon reformed but now all facing camera, the lead center in a frozen vogue pose one arm high, four dancers kneeling on one knee, five long shadows converging at the lens; micro push-in, hold. Loopable.

Performance rules: lead stoic, owns the horizon; dancers precise, faces neutral. Continuity: same five people, same wardrobe, same rooftop and light. Audio intent: every cut and formation change on a hit of
@Audio1
; faint foley only (wind, gravel, fabric). Quality bar: expensive editorial music video, natural cinematic grade, no AI gloss, no glow, no lens-flare abuse.

I just built my own coding harness, just like Claude Code. (100% open-source)

A harness is the code wrapped around an LLM. The model only decides the next step, and the harness handles everything else, planning, tools, memory, and safety.

In simple words model acts as the brain and the harness provides the hands.

A coding harness is that same wrapper pointed at a codebase. It turns a plain text generator into something that reads files, edits code, runs tests, and fixes real bugs.

Here's everything that went into mine:

- the core agent loop
- file tools that double as external memory
- planning for long-running tasks
- subagents that work in their own context
- sandboxed execution in a throwaway VM
- human-in-the-loop approval
- persistent memory and checkpointing

The whole thing is built on CrewAI, a 100% open-source framework.

I also wrote an article on this. It covers everything that goes into building a coding harness, the agent loop, planning, subagents, sandboxing, memory, and checkpointing, built step by step.

The article is quoted below.

If you want to become GOAT-level with AI agents, try these 12 free courses (save them):

离谱！打死我都想不到，有公司愿意把这种东西公开出来。

AI 视频圈最没用的就是"只放成片"，成片可以随便给人看，但工程文件是命根子。

Higgsfield 这次把 Originals 的工程文件全开源了：
每个镜头的完整 prompt、角色参考图、音频母带、生成参数，点开项目全能看。

我翻了一圈，学到了很多：
角色先做成一整套参考图锁一致性，每个镜头单独一个文件夹，哪个镜头坏了只重跑那一个。

看完你就明白了，大片是把 AI 当真剧组管出来的，写一条 prompt 等奇迹这种事从来不存在。

挑了个我最喜欢的镜头，提示词原文长这样：
Seedance 2.0 on Higgsfield AI Full open-sourced prompts & assets below:

Prompt:
Slow-motion handheld close-up, 16:9. Shaky organic handheld shot shoving in close on an older grey-haired man in a soaked white shirt and tie, chest-deep in churning green-grey flood water clutching a glass bottle, mouth open mid-shout — while all around him a dense crowd churns: a woman shoves past his shoulder, two men wrestle a floating chair behind him, someone falls backward into the water beside him sending up a slow sheet of spray, more panicking bodies wade and thrash in the background. In slow motion water explodes across the whole frame, droplets suspended everywhere. Dynamic jostling handheld movement at water level in the thick of the crowd. Sickly institutional palette — seafoam green, cream, muted teal, cold reflective water. All faces anatomically correct, stable, realistic, genuine emotion, no warping. Heavy coarse analog 35mm film grain baked into every frame, thick and consistent, grain crawling over highlights and shadows, like a scanned celluloid film print, never clean, never digital-looking. Flat overhead fluorescent light, dramatic slow-motion intensity, cinematic photorealistic, ultra-detailed.