#CVPR2026 GR3D 🧊 — A single VLM that grounds in 2D, grounds in 3D, and reasons with visual chain-of-thought — all at once.
Excited to share our paper, Grounded 3D-Aware Spatial Vision-Language Modeling!
Someone built an AI agent that searches Reddit, X, YouTube, HN, TikTok, Polymarket, and the web in parallel.
Scores everything by real upvotes, real likes, and real money.
Synthesizes it into one brief. In seconds.
It's called /last30days. 28,700+ stars on GitHub.
You type one command. The agent fans out across every platform at once. Reddit threads. X posts. YouTube transcripts. Polymarket odds backed by actual money. HN comments. GitHub commits. It scores each source by what real people engaged with. An AI judge synthesizes the whole thing into one grounded summary of the last 30 days.
Here's what it does:
→ Searches Reddit for top upvoted threads and comments on any topic, person, or company.
→ Pulls X posts and scores them by likes and recency. Not algorithmic feed. Raw signal.
→ Transcribes and searches YouTube videos. Finds what was actually said, not just the title.
→ Reads TikTok engagement. Surfaces what creators and communities are actually talking about.
→ Queries Polymarket odds. Real money bet by real people on what happens next.
→ Searches Hacker News. The technical community's unfiltered take.
→ Searches GitHub commits and PRs. What someone is actually shipping right now.
→ Runs all sources in parallel. Scores them against each other by engagement weight.
→ AI agent judge synthesizes everything into one brief. No raw dump. A grounded summary.
→ Zero config to start. Reddit, HN, Polymarket, and GitHub work immediately.
→ One setup wizard unlocks X, YouTube, TikTok, and more in 30 seconds.
→ Installs into Claude Code, Codex, Cursor, Copilot, Gemini CLI, and 50+ agent hosts.
Here's the wildest part:
Google doesn't touch Reddit comments or X posts. ChatGPT has a Reddit deal but can't search X or TikTok. Gemini has YouTube but not Reddit. Claude has none of them natively. Every platform is a walled garden with its own API, its own tokens, its own auth.
No single AI has access to all of it. Until you bring your own keys and bridge them with an agent.
That's the unlock. Not one better search engine. A dozen disconnected platforms, scored against each other by what real people actually engaged with and bet real money on.
Google aggregates editors. /last30days searches people.
Perplexity Pro: $20/month. $240/year.
ChatGPT Plus: $20/month. $240/year.
You(dot)com Pro: $15/month. $180/year.
/last30days: $0. Unlimited queries. Unlimited topics. Your API keys. Your agent. Forever.
28,700+ stars. 2,431 forks. MIT licensed.
MIT licensed. Self-hosted. Open protocol. Free forever.
100% Open Source.
Github repo: https://t.co/jSgEaiye2H
supervision just hit 40,000 GitHub stars!
it now powers over 6.5k open-source computer vision projects, including all my demos like basketball AI
link: https://t.co/xXMRaS4ejS
WAIT. This is actually insane.
A senior dev dropped the SOUL .md template behind his Hermes Agent. Says he's never shared this before.
The sections that turn your agent from a chatbot into an autonomous operator:
→ Stance: direct, opinionated, push back when I'm vague
→ Accountability: surface opportunities, flag stalled loops
→ Autonomy: broad freedom except for irreversible actions
→ Mission: priorities, active builds, debt, sunset candidates
→ Pushback: disagree openly, earn it with evidence
→ Operating Mode: orchestration, not solo execution
The author says three sections decide if the agent acts like an operator: Stance, Autonomy, and Mission.
The Autonomy section alone is worth the whole template. Most builders never write this out and then wonder why their agent asks permission for every action.
(Full template in the comments)
A simple MPC controller for path tracking implemented in python.
Few resources actually teach it without assuming a PhD first.
This open-source project does exactly that:
An iterative MPC tracking controller (built with CVXPY and MuJoCo) that shows how to make a robot (or autonomous vehicle) follow a path and avoid obstacles in real time.
→ Autonomous logistics, robot navigation, vehicle control; there's a path-following algorithm behind all of it.
→ Most implementations are black boxes. This one shows you what's actually happening.
→ It uses convex optimization (no non-linear solver), and still handles real vehicle physics through iterative linearization.
What makes it practical is that you can run it with the mushr car model in MuJoCo, or headless without any physics simulation if you just want to understand the logic.
If you're making the jump from classical control to real-time optimizatio, this is a very good starting point.
📌 Repository: https://t.co/dzi784Vgmq
——-
Weekly robotics and AI insights.
Subscribe free: https://t.co/9Nm01QUcw3
The Library of Alexandria created the first catalog of all human knowledge 2,300 years ago, and a team of fewer than 20 people just finished the modern version and made it free for the entire planet.
It is called OpenAlex. The name is not an accident.
The ancient library had the Pinakes, a catalog mapping every scroll, every author, every subject. When the library fell, the map of what humanity knew fell with it.
For the last two decades, that map existed again, but it was locked up.
Elsevier owns Scopus. Clarivate owns Web of Science. If your university could not afford the subscription, you could not see the structure of science itself. Entire countries were priced out of knowing what research existed.
OpenAlex indexes 474 million scholarly works. Every author disambiguated. Every citation traced. Every institution and funder connected. It updates with roughly 50,000 new works every day.
The whole thing is CC0. Not just free to search. Free to download, copy, sell, and build on. The API allows 100,000 requests a day without an account.
The ancient library burned and the catalog was lost for two millennia.
The new one cannot burn. Anyone can hold a copy.
https://t.co/peUYYpucnc
🚨 BREAKING: Google Gemini can now completely edit your videos for you so you can create polished content at lightning speed.
Here are 10 Gemini prompts to edit videos 10x faster: 👇
I'm dead. I don’t even know if I'm laughing because of the drinks or because you're hilarious 🤣🍸
🩷 Mckenna Grace
🩶 Natalie Portman
🤍 Dakota Johnson
Nano Banana 2 🍌via Gemini App
Do you like Super Mario video game?!🎮 Let's make it more attractive!🔥
Iggy Azalea, Sophie Thatcher, Kylie Jenner, Sophie Rain ❤
👉🏻Subscribe for more content!⚡
Nano Banana Pro via Hailuo AI
Prompt:
{
"subject": "Iggy Azalea lookalike",
"hair": {
"color": "blonde",
"length": "long, reaches past mid-back",
"style": "voluminous, wavy, natural flow with soft movement",
"details": "shiny, healthy, slightly tousled"
},
"face": {
"expression": "direct eye contact, confident and slightly seductive, lips gently parted",
"makeup": "glossy pink lips, heavy black winged eyeliner, long false lashes, defined arched brows, subtle blush and contour, dewy skin finish",
},
"body": {
"build": "curvy, hourglass figure, large bust",
"pose": "standing, torso leaned slightly forward toward camera, weight on one leg, left hand gripping the side of her dress near the hip, right hand cradling the plush toy",
"skin": "fair, smooth, with realistic subsurface scattering and natural highlights from side lighting"
}
},
"clothing": {
"dress": {
"type": "shiny satin or latex-style mini dress",
"color": "vibrant hot red",
"neckline": "deep plunging V-neck with thin spaghetti straps",
"trim": "white lace ruffles along neckline and hem",
"skirt": "short and flared with layered white lace petticoat peeking out and a small sign 'Keor' on border",
"fit": "form-fitting bodice, accentuates bust and waist",
"material": "glossy, reflective satin-like fabric with realistic sheen and folds"
},
"accessories": [
{
"type": "choker",
"description": "pink lace choker with delicate pattern"
},
{
"type": "stockings",
"description": "white floral lace thigh-high stockings with intricate pattern"
}
]
},
"props": [
{
"name": "plush_toy",
"description": "round green and white plush toy with simple cartoon face and eyes, resembling a green mushroom from Super Mario series",
"color": "bright green upper half, off-white/cream lower half",
"position": "held with both hands in front of her lower chest / stomach area, right hand supporting from below",
"size": "medium plush, roughly the size of a large grapefruit"
},
{
"name": "white_platform_boots",
"description": "tall white platform boots with pink laces/ribbons and decorative elements",
"position": "standing upright on a white surface/shelf in the right background"
},
{
"name": "mini_crown",
"description": "small ornate golden crown/tiara with red gemstone in the center",
"position": "placed on the surface next to the white boots"
}
],
"background": {
"setting": "cozy indoor room with light beige/pinkish walls, gaming or cosplay aesthetic",
"left_wall": {
"element": "framed poster",
"description": "Mario character poster",
},
"neon_lighting": {
"main_neon": "pink neon sign on wall with partially visible glowing text "Princess Peach",
},
"right_side": "pair of white platform boots and small golden crown on a white shelf/surface, fluffy white feather wings or decorative elements",
"overall_atmosphere": "bright, playful, slightly seductive cosplay vibe with Mario/Princess Peach theming"
},
"lighting": {
"primary": "soft natural daylight from front-left, creating strong rim lighting on hair and dress, bright highlights on skin and glossy surfaces",
"secondary": "pink neon glow adding vibrant color accents and soft fill light",
"shadows": "defined but soft shadows on the right side of the subject and background",
"mood": "bright, high-key, cinematic with playful and alluring atmosphere"
},
"camera": {
"angle": "slightly low angle, eye-level to subject",
"framing": "medium shot from upper thighs to head",
Bending over the desk, writing your name on my very naughty list. Come find out what happens when you make the list 📝🫦💦
Elizabeth Olsen
Cailee Spaeny
Emily Rudd
Freya Allan
Prompt:
{
"subject": {
"Emily Rudd lookalike",
"build": "hourglass figure, large bust",
"hair": {
"color": "blonde",
"length": "very long, extending well past the shoulders and mid-back",
"style": "voluminous soft waves with center part, heavy volume on top, hair flowing dramatically over the right shoulder and cascading down the back",
"texture": "shiny, healthy, silky waves"
},
"face": {
"eyes": "eyes with long lashes",
"lips": "full, glossy with soft pink-nude lipstick",
"makeup": "natural flawless base, subtle contouring, defined eyes with mascara, glossy lips",
"glasses": "glasses with big black frames",
"expression": "seductive and confident, direct intense eye contact with viewer, slight head tilt, parted lips"
},
"skin": "fair, smooth, flawless complexion with soft glow"
},
"clothing_and_accessories": {
"top": "black strapless bandeau-style fitted top, smooth matte fabric, revealing shoulders and upper chest",
"skirt": "black short mini skirt with structured fit and subtle ruffled/layered hem detail at the bottom",
"belt": "wide black utility/tactical-style belt with large prominent silver metal rectangular buckle and small attached pouches on the side",
"stockings": "black sheer thigh-high stockings with subtle sheen",
"footwear": "black glossy patent leather knee-high stiletto boots with pointed toes and high heels",
"other": "small silver hoop earring visible in left ear"
},
"pose_and_action": {
"overall_pose": "bent forward at the waist, leaning over a desk in a dynamic three-quarter view, upper body angled toward the viewer",
"torso": "strongly arched and tilted forward",
"legs": "kneeling or seated position with one knee raised, creating a strong curve from hip to knee",
"right_hand": "holding a blue wooden pencil, actively writing on a white sheet of paper on the desk",
"left_hand": "flat on the desk surface for support, fingers relaxed",
"head_and_gaze": "head turned toward camera, looking directly at viewer with seductive expression"
},
"environment": {
"setting": "modern minimalist office interior",
"desk": "dark wood desk with black top surface",
"chair": "black mesh ergonomic office chair positioned behind the subject",
"background_wall": "plain light gray wall",
"clock": "large round black wall clock with white face, black Roman numerals, and black hands showing approximately 10:10",
"shelf": "thin white floating shelf on the wall holding a small transparent glass spherical decorative object",
"desk_accessories": "black mesh pen/pencil holder containing colorful markers and pens on the right side of the desk"
},
"lighting": {
"type": "soft, bright, diffused indoor studio lighting",
"quality": "even and flattering with gentle highlights on skin and hair, minimal harsh shadows, professional photography lighting",
"mood": "bright, clean, slightly dramatic"
},
"camera_and_composition": {
"shot_type": "medium shot / medium close-up",
"angle": "slightly low angle from desk level, three-quarter view",
"framing": "subject dominates the frame from upper thighs to head, desk occupies lower third",
"depth_of_field": "sharp focus on subject, soft background",
"aspect_ratio": "portrait orientation"
},
"style_and_quality": {
"style": "photorealistic, hyper-detailed, commercial fashion photography",
"aesthetic": "sexy professional 'office siren' / seductive secretary vibe with tactical belt detail",
"mood": "confident, alluring, playful, powerful",
"technical_quality": "8K resolution, extremely sharp details on hair, fabric textures, skin, and reflections on boots"
},
ONE OF THE MINDS BEHIND JAVA GAVE A LECTURE WHERE HE BANNED HIMSELF FROM USING ANY BIG WORD UNTIL HE DEFINED IT FIRST AND IN DOING SO QUIETLY EXPLAINED HOW EVERY GREAT SYSTEM GETS BUILT
A talk from Guy Steele -- co-author of the Java spec and a designer of Scheme -- where the form of the talk is the lesson.
-> The moment you catch what he's doing, it rewires you. He starts with only the smallest words, then builds every larger idea live, in front of you, from the pieces he already gave you. The talk grows its own vocabulary as it goes.
That's the whole secret of language and system design he's smuggling in: you don't ship something huge and finished. You ship a small core and the means to grow it. Miss that and you spend your career fighting your own tools.
Memorizing a language was never the skill -> understanding how a good one is meant to grow is. And as AI starts generating the primitives you build on, knowing what makes a foundation extensible instead of brittle is the entire game.
Twenty-five years on, language designers still point to this talk as the cleanest demonstration of the idea ever performed.
Bookmark it & Watch how he builds the whole thing from nothing ↓
May Results 📈
50-21 (70.4%) Win Rate ✅
Every play was posted here or on the Free Page in bio.
And the best part? Even bigger results are on the way with the FIFA World Cup kicking off. ⚽🔥
This 1-hour Stanford lecture on "Sports Betting Math" - by the founder of a $160M sports platform - exposes the exact math books use to print millions off sports.
Bookmark it. Give it 1 hour today, no matter what. Most productive way to start your week.
Then read article below
Anthropic engineer:
"You're not supposed to prompt Claude. You're supposed to build a system that prompts itself."
this is one of the best workflows I've seen in a long time
in this video she breaks down exactly how most people are using Claude:
- the 14% you lose to CLAUDE.md before typing a word
- the automation workflows most users don't know exist
- the daily task pipelines that run without touching the keyboard
- the daily workflows Anthropic's own engineers automated first
if you've been using Claude for more than a month and never left the chat window, you've been using one agent when you could be running a team of them
instead of another show tonight, watch this
make sure to bookmark it before it gets lost in your feed
the guide is in the article below