I made it for fun, but I'd love suggestions on how I can improve it in any way. Thanks, love you.
I'm Adriano, one of the makers of Vaultrice. I'm excited (and a little nervous!) to share what we've been building.
For years, we found ourselves in a frustrating loop: whenever we needed a simple real-time feature—like a "who's online" list, a collaborative app, or just sharing state between a marketing site and our main app — we'd end up spending days setting up the same stack or discarded to do it. Setting it up, it always involved wiring together a database, a WebSocket server, an API, and managing the connection state. It felt like massive overkill for what we were trying to achieve.
We wanted a tool that felt as simple to use as the browser's `localStorage` API but worked across domains and devices, with real-time sync and security built-in.
So, we built Vaultrice.
It’s a key-value data store built on top of Cloudflare's Durable Objects, which gives you a strongly consistent backend for each data object. You interact with it through our TS/JS SDK, which comes in two flavors:
1. `NonLocalStorage`: A low-level client with a `localStorage`-like API (`setItem`, `getItem`, etc.) plus real-time events and presence (`.on()`, `.join()`).
2. `SyncObject`: A higher-level, reactive JavaScript Proxy. You just change a property on an object (`doc.title = 'New Title'`), and it automatically syncs to all other connected clients.
The goal is to let you build the real-time features you want in minutes, not days. We've also put a lot of thought into a layered security model, allowing you to go from simple API key restrictions all the way to server-signed object IDs and client-side E2EE.
We’ve just launched and would be grateful for any feedback from the HN community. What do you think of the API design? Are there use cases we haven't considered? Any thoughts on the security model?
We'll be here (or via email ([email protected])) to answer any questions. Thanks for checking it out!
Define your schema → attach callbacks → push streaming tokens → get structured events immediately.
We built it to solve the common problem we and our customers had: engineers spend more time on AI ops and plumbing than on actual feature development. This was often due to the challenge of using Python for I/O-bound, high-concurrency workloads and complexity maintaining pipelines with streams that use always-changing ML models.
Our solution is a high-performance runtime written in C++ with the core idea of defining AI logic as graphs. For instance, a basic voice-to-voice agent consists of STT → LLM → TTS nodes, while the connecting edges stream data and enforce conditions. This graph engine is portable (Linux, Windows, macOS) and can run on-device.
We built a few key features on top of this C++ core:
- Extensions. Runtime architecture decouples graph definition from implementation. If a pre-built component doesn't exist, you can register your own custom node/code and reuse it in any graph without writing any glue code.
- Routers. You can dynamically select models/settings on the per-node basis depending on the traffic as well as configure policies for fallbacks and retries to get the app ready for production.
- The Portal. A web-based control plane UI to deploy graphs, push config changes instantly, run A/B tests on live traffic, and monitor your app with logs, traces, and metrics.
- Unified API. Use our optimized models or route to providers like OpenAI, Anthropic, and Google through a single, consistent interface and one API key.
We have a Node.js SDK out now, with Python, Unity, Unreal, and native C++ coming soon. We plan to open-source the SDKs, starting with Node.js.
The docs are here: https://docs.inworld.ai/docs/runtime/overview
We're eager for feedback from fellow engineers and builders. What do you think?
Songs are written in plain, 7-bit safe text. Every line makes a voice/channel.
The examples given on the page should hopefully illustrate every feature, but as a quick overview: Sounds are specified using Anglo-style note names, with flat (black) keys being the lowercase version of the white key above so as to maintain one character per note. Hence, a full chromatic scale is AbBCdDeEFgGa. Every note name is interpreted as the closest instance of that note to the preceding one. +- skips up or down an octave, ~ holds the previous note for a beat, . skips a beat, 01234 chooses one of 5 preset timbres, <> makes beats slower or faster (for all channels), () makes the current channel louder or quieter. All other characters are ignored.
If you come up with a good tune, please share it in the comments!
I was juggling work and personal GitHub accounts with separate PATs for a long time and constantly forgetting to switch between them. Needed a way to commit to personal and work projects without the mental overhead of managing two Git identities.
My issue:
```
cd ~/work/important-project
git push
# Authentication failed - using personal PAT for work repo
```
Then the dance:
```
git config user.email "[email protected]"
# Update Git credential helper or remember which PAT to use
# Rinse and repeat every time I switch contexts
```
My solution (I'm sure others exist?)
```
# One-time setup
gitego add work --name "John Doe" --email "[email protected]" --pat "ghp_work_token"
gitego add personal --name "John" --email "[email protected]" --pat "ghp_personal_token"
gitego auto ~/work/ work
gitego auto ~/personal/ personal
# Now it just works cd ~/work/any-project
git commit -m "fix bug" && git push # Uses work identity + PAT automatically
cd ~/personal/side-project
git commit -m "new feature" && git push # Uses personal identity PAT automatically
```
How It Works
- Uses Git's native `includeIf` for identity switching
- Acts as a Git credential helper for automatic PAT selection
- Stores PATs securely in your OS keychain
- Single Go binary, works on macOS/Windows/Linux
No more context switching overhead. Just cd and commit.
GitHub: https://github.com/bgreenwell/gitego
Install: go install github.com/bgreenwell/gitego@latest
Feedback welcome!
Keep in mind, I built this as a personal tool, making it public in case others have the similar problems and can benefit from the solution!
The past two years have been a whirlwind. AI models now create images, videos, and even code faster than humans. I felt both amazed… and scared. If AI could do my job, what would be left for me?
Instead of waiting for the wave to wash over me, I decided to build something of my own. Video tools were too costly, so I started with images — the foundation of video — and began working on photor.ai, an AI-powered image generation and editing tool.
No design skills. No frontend experience. No budget.
1.V0 for the product page
2.Cursor for both frontend & backend code
3.GitHub for source control
4.Cloudflare Pages for hosting frontend
5.A tiny cloud server for the backend
It wasn’t smooth:
1.Broke deployments at 2 a.m. more than once
2.Rebuilt features from scratch when early versions failed
3.Hit API rate limits just hours before launch
There was a night I nearly quit. I told myself, “Maybe I’m not cut out for this. I’m just a CRUD dev.”
The next morning, I decided to keep going anyway. One bug at a time, one feature at a time, one tiny improvement at a time. Slowly, photor.ai came to life.
If there’s one thing I’ve learned, it’s this: you don’t need to be a “full-stack rockstar” to launch something in the AI era. Curiosity and persistence matter more than credentials.
If you’ve been hesitating, start now. Even if it’s messy, even if it’s small — the best way to understand this new era is to create in it.
I built a "platform" with tutorials that are spoon-feeding knowledge to people via panels that you are advancing by clicking spacebar or tapping. To make it different, I also wrote them with very "light" language and added few cringe jokes and elements.
Right now just to test the idea I added two tutorials: - Python Type Hints - Coding Interview Tips
Right now I am looking for feedback because I want to find out if this way of learning could be actually useful for anyone. Or if it's another idea of mine that fits into the category "cool, but no one wants that".
I will be really grateful for any feedback! Thank you!
I'm excited to share a small side project I've been working on: *www.chivalrytest.online*. This is a fun, quick online quiz designed to help people explore a concept that's often seen as archaic but holds timeless values: chivalry.
The idea behind this project isn't to be a historical authority. Instead, it's a lighthearted way to reflect on some core virtues that are still relevant today, such as:
* *Courage:* How do you react when faced with a difficult choice? * *Honor:* What role do integrity and responsibility play in your decisions? * *Humility:* Are you willing to admit your mistakes and learn from them?
The test asks a series of simple questions and provides a result that categorizes you as a specific type of knight, like a "Knight of Honor" or a "Knight of Wisdom." It's a quick and engaging way to spark some self-reflection.
I built this project using `html、JQuery、TailwindCSS` and it was a great learning experience. I'd love to get your feedback on the questions, the results, and the overall concept.
What do you think? Are these virtues still relevant in today's world?
Thanks for checking it out!
HOW DOES IT WORK 1. Upload a Loom or equivalent video where you describe what you want to be tested in your app 2. Gemini extracts test scenarios 3. We spawn multiple computer-use agents in parallel to explore your app and write Playwright code.
WHAT’S DIFFERENT Your voice and a screen recording is the most natural way of describing what you want to be tested in an app. Compared to other test builders (AI powered or not) you do not have to spell out steps, think about test design or dynamic data within the app.
HOW YOU CAN TRY IT We have a no-sign-up demo at https://app.fumedev.com/demo
Would love to hear what you think of this UX for creating Playwright tests. Especially if you wrote E2E tests before or used any of the test builders.
I'm really excited to release Claude Control today.
You can check it out here - https://claudecontrol.com
It's an app which connects claude code to your Slack & Discord and lets you open PRs and iterate on them without leaving the chat app.
More importantly, it lets you share this with your whole team or community.
Quite frequently, non-technical people on the team or support engineers would like to understand how something in the product works, so this lets them get their answer without setting up anything, just by chatting with Claude Code.
I think it also unlocks a lot of other capabilities, a few that come to mind: - run it in your docs repo & let it answer questions for your slack/discord community - connect notion via mcp and let it work as an always available knowledge base in your slack
There's probably a lot more interesting use-cases.
It's currently free in open beta and I can't wait for you to try it out. I'm actively iterating on it and looking for some engaged early users to help me shape the app direction.
Hope you like it & find it useful!
Key features: • Automatically discovers hosts from your ~/.ssh/config • Real-time CPU, memory, disk, and GPU monitoring • Timeline charts showing historical usage • One-line installer: curl -sSL https://raw.githubusercontent.com/tsugumi-sys/ssh-monitor/ma... | bash • SQLite storage for historical data • Cross-platform (Linux, macOS Intel/ARM)
GitHub: https://github.com/tsugumi-sys/ssh-monitor
Would love feedback on the UX and feature suggestions!
For example:
1. Generated mind maps cannot be edited.
2. Mind maps cannot be exported to other formats.
My idea was to create a lightweight tool that would meet the needs of novice users and require only a single operation. Therefore, I've broken down various functions into separate software, including PDF, image, URL, and text generation. Other features will be added later.
I made a tiny tool called mirroranimage.com. It does just one thing – flip or mirror any image – but it does it fast and entirely in your browser.
Why I built it I often needed to mirror images for quick design checks, social posts, or just to see how something looks reversed. The problem? Most tools are slow, ad-filled, or upload your images somewhere. I wanted something instant, private, and pleasant to use.
What you can do: • Flip horizontally or vertically with one click • Drag & drop or paste directly from clipboard • No sign-up, no upload, no tracking – works 100% locally • Download instantly when you’re done
Why it’s nice: • It’s fast. Like, under 3 seconds fast. • No clutter – just you and your image • Works on desktop & mobile
Try it here → mirroranimage.com
I’d love feedback on what could make it even smoother. Would batch mirroring or rotation be useful? Let me know!