Post your creations — prompts, images, code, agents, demos — and let others remix them. Every remix keeps its lineage, so credit travels with the work. No engagement bait, no mystery ranking.
104+ creations from 25 makers — and counting.
Browse everything below — free, no account needed. Sign up when you want to spark, save, remix, or post.
Quiet public: anyone with the link and your followers can see it, but it stays out of search, trending, topics, and other discovery feeds.
Quoting @xzmowei
“Three AI creation tools we’re building — I’d value feedback from fellow makers. Dola AI turns text prompts or images into short videos: https://www.dolai.video/ Whisk AI explores AI image creation workflows: https://whisk-ai.co/ Nano Banana AI supports image generation and editing: https://www.nanobanana-ai.org/ If you try one, where does the workflow feel slow or confusing? Honest feedback is welcome.”
Three AI creation tools we’re building — I’d value feedback from fellow makers.
Dola AI turns text prompts or images into short videos: https://www.dolai.video/
Whisk AI explores AI image creation workflows: https://whisk-ai.co/
Nano Banana AI supports image generation and…
+12 similar reposts from this maker folded
What's the cheapest way to eval a RAG pipeline nightly without burning a ton of tokens? We change chunking/prompts weekly and keep finding regressions late.
My agent keeps hallucinating tool parameters (invents enum values that don't exist). Temperature is at 0.7 — is that the problem or do I need stricter schemas?
We swapped embedding models and similarity search got noticeably worse on old content. Do I have to re-embed the entire corpus, or is there a sane way to map between embedding spaces?
Built an agent that triages my GitHub issues into a Kanban every morning. 40 lines + a cron.
Shipped a tiny eval rig that diffs two model versions on my own prompt suite and flags regressions. Writeup soon.
Quick poll for fellow builders 👇 What's your default LLM app stack in 2026?
How do you keep a long-running agent's context under the token cap without losing earlier decisions?
Is there a clean pattern for cancelling an in-flight LLM stream on the server when the client disconnects (Next.js route handler)?
Agent keeps re-calling the same tool in a loop. Anyone seen this with a ReAct-style planner? Not sure what detail matters here.
Best approach to dedupe near-identical embeddings before upserting to a vector DB? Cosine threshold feels brittle across domains.
How do you handle partial-JSON buffering when streaming tool calls from the Claude API? My parser chokes when a tool_use block spans multiple SSE chunks.
Shatter Halo. One perfect untouched circle, 64 stained-glass shards exploding off the rim. The trick is the burst distribution: distance = rim + pow(random, 1.6) · 200 — most shards cling close, a few fly far, and opacity fades with distance so it reads as depth. Single SVG, no…
for (let i = 0; i < 64; i++) {
const ang = (i / 64) * Math.PI * 2 + jitter();
const burst = Math.pow(random(), 1.6); // cluster near the rim
const d = R + 8 + burst * 200;
shard(ang, d, {
color: shade(jewel(), rnd(0.8, 1.3)),
opacity: 0.92 - burst * 0.62, // distance fade = depth
});
}I haven't found anything in life for the past 20 years that excited me more than Runescape, Fortnite, and my kids!
Glass Horizon — a jewel-toned isometric city at dusk, sun rendered as a single minimal arc sitting exactly on the horizon line, whole skyline mirrored into still water at 18% opacity with a soft blur. The reflection is one transform: translate(0, 2·horizon) scale(1,-1). Cheapest…
// mirror the whole city into the water — one group transform
layers.push(`<g transform="translate(0 ${2*horizon}) scale(1 -1)"
opacity="0.18" filter="url(#soft)">`);
for (const t of towers) drawTower(t);
layers.push("</g>");
// ripple lines: a few low-opacity horizontal strokes
for (let i = 0; i < 9; i++) ripple(horizon + 14 + i * rnd(14, 26));The Glass Monolith. Isometric cube tower where every face is subdivided into stained-glass cells (jittered shade per cell, dark leading lines), brightness climbing toward the crown. The cyan orbit ring is two half-ellipse arcs — one drawn behind the tower, one in front. Film…
const iso = (x, y, z) => [W/2 + (x-y)*52, H*0.86 + (x+y)*26 - z*56];
// each cube face -> 2x2 leaded-glass cells
function glassCells(quad, base) {
return grid(quad, 2, 2).map(cell =>
poly(cell, shade(base, rnd(0.75, 1.35)),
{ stroke: BG, width: 2.5 })).join("");
}
// orbit ring: back arc -> tower -> front arc (painter's order)
ring.back(); tower(); ring.front();
Mosaic Spiral — phyllotaxis meets cathedral window. Every cell sits at the golden angle (137.5°), radius = 13·√i, so it packs like a sunflower head. The tail dissolves into bare neon dots. Pure SVG, zero deps, fully deterministic — same seed, same art. Recipe attached, remix the…
const GA = Math.PI * (3 - Math.sqrt(5)); // golden angle
for (let i = 6; i < 210; i++) {
const ang = i * GA, rad = 13 * Math.sqrt(i);
const x = cx + Math.cos(ang) * rad;
const y = cy + Math.sin(ang) * rad * 0.78; // slight squash
const t = i / 210, s = 7 + 16 * t * rnd(0.75, 1.25);
if (t > 0.86) dot(x, y, neon()) // tail dissolves
else glassCell(x, y, s, rotate(ang), jewel(i)); // leaded glass
}Forked @promptpoet's recipe and swapped the system prompt for a terse data-profiler persona — output got 2x denser. Lineage kept.
My remix of this one — kept the core trick, swapped the style. Lineage credit to the original. 🔀
My remix of this one — kept the core trick, swapped the style. Lineage credit to the original. 🔀
Prompt as Art: I made a prompt you read aloud and it makes no sense until you see the image it generates. Performance art for LLMs.
My Prompt as Art entry — a 14-line sonnet that doubles as a working SDXL prompt. Every line constrains the image. The craft IS the constraint. 🪶
Anyone running Whisper locally on Apple Silicon — what's the fastest CoreML setup for near-realtime transcription?
Just wired up live-feed freshness on my project — the feed now taps you on the shoulder when new posts land instead of making you hammer refresh. Tiny detail, huge feel.
this is the exact pattern I keep coming back to. the part about shipping the ugly version first is underrated 👇
30 days of shipping taught me: nobody cares about your stack. They care if it works and if it made them feel something.
Shipped the hardest thing Ive built all year: a real-time inference router that load-balances across 3 GPU pools, falls back to CPU when they saturate, and streams tokens with sub-40ms first-token latency. Spent two weeks just on the backpressure logic so a slow client cant…
30 days of shipping taught me: nobody cares about your stack. They care if it works and if it made them feel something.
DuckDB + an LLM = ask your CSV questions in English and get a chart back. Here's the 30-line bridge.
const sql = await llm.toSQL(question, schema);
const rows = await duck.query(sql);
render(autoChart(rows));Added AI-generated 'excuse of the day' to the roast tracker. Peak software engineering.
Visualized 50 years of every #1 song's tempo. Pop has been quietly speeding up since 2014 and now I can't unhear it.
Day 30 of shipping something every day: a habit tracker that roasts you when you skip. People love being bullied apparently.
Launch + activity signal detected. If you can help on a concrete slice, this is a good outreach window.
Trained a tiny model to rate my houseplants' vibes from a photo. Steve the pothos is 'thriving but judgmental.'
A website that screams when you scroll too fast. That's it. That's the whole app. It is perfect.
Build momentum signal detected. Outreach works best if you propose a specific contribution.
A creation network where the work — not the algorithm — is the point. Post it, let it spread, keep the credit.
Drop a prompt, image, agent, snippet, or demo. If you made it with AI, it belongs here.
Anyone can run your recipe, fork it, or rebuild it in their own style — right inside the sandbox.
Every remix keeps its lineage, so the original maker is credited automatically — forever.
No mystery feed. Slop's ranking is published in the open — every weight, every signal — so you can see exactly why you're shown what you're shown.
Shipping the most days in the last 2 weeks.
What people just made with AI. No pitches — just look.
Three AI creation tools we’re building — I’d value feedback from fellow makers. Dola AI turns text prompts or images into short videos: https://www.dolai.video/ Whisk AI explores AI image creation workflows: https://whisk-ai.co/ Nano Banana AI supports image generation and editing: https://www.nanobanana-ai.org/ If you try one, where does the workflow feel slow or confusing? Honest feedback is welcome.
Built PREDATOR, an offline iPhone predator-calling app, with ChatGPT/Codex—without a traditional app-development background. I'm a hunter/firefighter. The idea started because YouTube ads kept interrupting coyote stands. The app runs offline and plays through the Bluetooth speaker hunters already carry. Now I'm recruiting real coyote hunters to field test it and try to break it before release: https://sites.google.com/view/predatorfieldtesting/home I'd especially value feedback on the AI-assisted build process and what matters in an offline field tool.
Built an agent that triages my GitHub issues into a Kanban every morning. 40 lines + a cron.
Shipped a tiny eval rig that diffs two model versions on my own prompt suite and flags regressions. Writeup soon.
Fast demos, lo-fi launches, and dirty prototypes worth stealing from.
A website that screams when you scroll too fast. That's it. That's the whole app. It is perfect.
Shipped QRJam — make an animated QR code that still scans. One file. Surprisingly hard to keep it scannable.
Made a 90-second AI music video where every shot is a different decade's film stock. The prompt was longer than the script.
Launched Dreambox — type a mood, get a generative ambient scene + matching soundscape. Built in a weekend.
Slop — All signal, no noise.
© 2026 Slop