Open Roles (REMOTE, non US):

Animators (2D/3D or hybrid—experience with AI tools a plus) Writers Programming interns

Why join us?

Fully remote, async-friendly Ship real things that impact how kids learn Fast, creative, zero-corporate-BS environment

Direct access to founders; real ownership

Opportunity to shape our tools and shows from the ground up

To apply: Email jobs [at] studyturtle.com with your background, a few sentences about why edtech is important to you, any feedback on our episodes, and why this mission excites you. Please include links to any relevant work.

We are the hiring company and will reply to all genuine applicants. No recruiters, no agencies, please.

Animators (2D/3D or hybrid—experience with AI tools a plus)

Junior Software Developers (Python, Node, JS, or open to learning new tech)

Interns (all backgrounds, generalists, tech, or creative)

We are also looking for writers (different kind of hacking)

Important: You must be a parent (of any age child) We want people who care deeply about kids and learning. Lived experience as a parent is essential—our mission is to build things real families want.

Why join us?

Fully remote, async-friendly Ship real things that impact how kids learn Fast, creative, zero-corporate-BS environment

Direct access to founders; real ownership

Opportunity to shape our tools and shows from the ground up

To apply: Email jobs [at] studyturtle.com with your background, a few sentences about why edtech is important to you, and why this mission excites you. Please include links to any relevant work.

We are the hiring company and will reply to all genuine applicants. No recruiters, no agencies, please.

Deep Mind also did a paper with grid cells a while ago: https://deepmind.google/blog/navigating-with-grid-like-repre...

There are actually a lot of people trying to figure out spatial intelligence, but those groups are usually in neuroscience or computational neuroscience. Here is a summary paper I wrote discussing how the entorhinal cortex, grid cells, and coordinate transformation may be the key: https://arxiv.org/abs/2210.12068 All animals are able to transform coordinates in real time to navigate their world and humans have the most coordinate representations of any known living animal. I believe human level intelligence is knowing when and how to transform these coordinate systems to extract useful information. I wrote this before the huge LLM explosion and I still personally believe it is the path forward.

Open Roles (REMOTE, non US):

Animators (2D/3D or hybrid—experience with AI tools a plus)

Junior Software Developers (Python, Node, JS, or open to learning new tech)

Interns (all backgrounds, generalists, tech, or creative)

We are also looking for writers (different kind of hacking)

Important: You must be a parent (of any age child) We want people who care deeply about kids and learning. Lived experience as a parent is essential—our mission is to build things real families want.

Why join us?

Fully remote, async-friendly Ship real things that impact how kids learn Fast, creative, zero-corporate-BS environment

Direct access to founders; real ownership

Opportunity to shape our tools and shows from the ground up

To apply: Email jobs [at] studyturtle.com with your background, a few sentences about your kids, and why this mission excites you. Please include links to any relevant work.

We are the hiring company and will reply to all genuine applicants. No recruiters, no agencies, please.

GGBench is a benchmarking platform to compare graphics produced by AI—especially code that renders visuals (still and animated). You vote on side-by-side outputs, and we rank models with an ELO system (the same concept chess uses). It’s simple, transparent, and designed to push the whole field forward.

Why this benchmark needs to exist Most LLMs are fantastic at writing code, but they’re basically flying blind: they can’t actually see what they produce. They output p5.js or Three.js scripts and call it a day. Whether that code renders beautifully—or at all—has been tricky to evaluate at scale.

Meanwhile, video diffusion models chew through compute to produce every pixel of every frame. Cool tech, but expensive, slow to iterate, and hard to control. Code-rendered graphics (p5.js, Three.js, shaders, SVG) are the opposite: cheap, fast, and precise. If I want a circle at (100, 100) moving at 2 px/frame, I can guarantee it. Determinism and control matter.

GGBench is my way to compare that code-first approach across models, in public, with ELO-based rankings that’s fair and hard to game.

What we measure (and why) We’re testing an LLM’s ability to write code that renders:

Still images: composition, color, structure, style fidelity. Animations: motion quality, timing, easing, coherence across frames. Because this is code, we also get repeatability (same prompt, same seed, same output), plus fine-grained prompts that check understanding of geometry, physics-ish motion, layering, and programmatic art patterns.

Categories you can explore Early categories include:

Cityscape – skylines, parallax scrolling, night/day cycles Nature – particles, waves, noise fields, flocking Abstract – generative patterns, symmetry, tiling, shaders You can filter the leaderboard by category to see which models excel where. Some models are great at structured geometry; others shine with texture and noise.

Why start with p5.js (and what’s next) I love p5.js: it’s approachable, expressive, and perfect for rapid iteration. But GGBench isn’t stopping there. We’re adding:

Three.js for 3D scenes and camera choreography WebGL/Shaders for low-level effects and speed SVG for clean vector output and accessibility As we broaden render engines, the benchmark will reveal which models can generalize across 2D, 3D, and shader-based tasks.

What this unlocks for the community For developers: a clear target to optimize against, with transparent signals on what’s working. For researchers: a public dataset of prompts, code, outcomes, and votes for deeper analysis. For creators: a practical way to choose the right model for your aesthetic and your stack. Roadmap (near-term) Three.js categories and 3D animation tests Per-prompt and per-engine breakdowns Submission flow for new models and agents “Tournament mode” with fixed prompts and seeds API for automated match scheduling and result ingestion Jump in If you’re curious where today’s models stand—or you just enjoy generative art—come help shape the rankings.

Start Voting: pick your favorite in head-to-heads View Leaderboard: see which models are rising About: read how everything is calculated (and verify it yourself) This is GGBench’s first public step. It’s opinionated—code first, community first, transparency first—because that’s what I think will actually move AI graphics forward. If that resonates with you, I’d love your votes, your critiques, and your pull requests.

Would love any feedback on this!

Comments URL: https://news.ycombinator.com/item?id=44865221

Points: 1

# Comments: 0

Strict age calibration (matching phrasing and examples to each developmental level)

Concrete analogies (“volcanoes are like shaken soda bottles”) and kitchen-table experiments you can actually do

Distark is hiring, an edutainment brand building learning videos for Education. Our small, passionate team creates animated shows and learning tools that help kids (ages 3-9) fall in love with curiosity and real-world learning. We use custom automation to speed up everything from story writing to animation. See what we’re making: https://www.youtube.com/watch?v=c46VaM_VZGU An example of some of our software, ChatGPT for families: https://www.studyturtle.com/ask

Open Roles (REMOTE):

Animators (2D/3D or hybrid—experience with AI tools a plus)

Junior Software Developers (Python, Node, JS, or open to learning new tech)

Interns (all backgrounds, generalists, tech, or creative)

Important: You must be a parent (of any age child) We want people who care deeply about kids and learning. Lived experience as a parent is essential—our mission is to build things real families want.

Why join us?

Fully remote, async-friendly Ship real things that impact how kids learn Fast, creative, zero-corporate-BS environment

Direct access to founders; real ownership

Opportunity to shape our tools and shows from the ground up

To apply: Email jobs [at] studyturtle.com with your background, a few sentences about your kids, and why this mission excites you. Please include links to any relevant work.

We are the hiring company and will reply to all genuine applicants. No recruiters, no agencies, please.

The way I think about this tool is that it is way to spark and test our curiosity. Is the child really interested in space or just randomly asking questions. Tools like this can help you as parent determine this faster.

For family tools like these I would want parents to know everything the child is asking about.