PH热榜 | 2025-12-16

Hi PH! 👋 Makers here, launching Readever today. We built Readever because “AI for reading” often means summaries after the fact, but the real pain is getting stuck while reading. Readever helps you read inside the text:

1. In-Context Q&A: Highlight any sentence while reading and ask questions where you’re stuck without leaving the page.
2. Proactive Reading Guidance: It adapts to your goals and level, proactively showing Highlight Cards so you get help even when you don’t know what to ask.
3. 5,000+ AI Reading Mentors: Read with thinkers, founders, writers, and historians. Ask them questions or let mentors debate to reveal different perspectives.
4. Built-in Translation: Read across languages without breaking your flow.
5. AI Book-Finding Agent: Describe what you want in natural language and get book recommendations tailored to your taste and intent. Readever is your next Knowledge Curator.
6. Memory system: Readever keeps you inside the text with fewer interruptions and faster comprehension, from the first page to the next book.

Readever isn’t for passive reading, it’s for understanding.
And yes, it’s totally free on the web!!!

The idea seems to quite new and creative for me. I really like to try and see how it works. Thanks for your work!

Is it possible to use an abstraction of a famous person's voice? How does that work if there's a monetisation aspect to it?

Great idea and a great product! Is there an option to read the book with the books author?

I liked the concept, but the ux needs a bit more work, It's very hard to understand what's going on and the use of already made comments on books.

I like the focus on visibility and control. If something is running on my Mac, I want to know about it no exception.

cool. AI highlighted some patterns I hadn’t noticed. How can I customize or choose my own character ?

Cool! Amazing for reading!

Congrats! Can users upload their own books, and does personalization still work with imported titles?

How does Readever personalize the reading journey for each user, especially readers with different goals?

I used it last night and genuinely felt accompanied while reading . a surprisingly warm experience.

"The 'dinner party' concept is such a cool hook. 🤯 I'm really curious to see how the AI mentors debate each other—does the AI actually mimic their specific rhetorical styles? Definitely trying this out. It makes reading feel less lonely."

What role do AI companions play during reading? Are they more like mentors, analysts, or reading buddies?

How large is the book catalog, and how does the public library integration function for users?

Congratulations on the launch of Readever! Honestly, your product positioning really caught my eye—it’s a rare and genuinely innovative product. I’ve already thought of many fun and interesting ways to use it, like having Jobs read his own biography with me and then asking him how he felt at different moments in his life. Keep it up!

I used it for fiction and loved how the AI companion picked up emotional shifts I totally missed.

How do you ensure accuracy when the AI comments on complex topics like economics, philosophy, or science?

Would Readever be useful for university students, researchers, or people writing a thesis?

From a product and AI perspective, how do you balance maintaining a consistent ‘persona voice’ while still adapting insights to different reading styles and genres?

Congrats on the launch! I'm curious about the workflow after reading, do you currently support exporting highlights or book notes to Notion or syncing with Readwise?

The memory system remembering my earlier confusion was surprisingly helpful.

How is Readever fundamentally different from NotebookLM, Readwise, or other reading/summarizer tools?

Really nice work, this looks clean and well done.

Just curious, have you ever worked with social media influencers to help show people how the software works and get more eyes on it?

Bro your launch video is something 😅😅

How does the system decide which concepts need explanation and which don’t?

Uploading my own book worked smoothly, and the annotations still felt tailored.

Short book summaries with narrated voices of famous personas could be a great addition)

How do you prevent the AI from overwhelming readers with too many notes or interruptions?

Hey Product Hunt — I’m Zack Li, CTO and co-founder of Nexa AI 👋

We built NexaSDK for Mobile after watching too many mobile app development teams hit the same wall: the best AI experiences want to use your users’ real context (notes, photos, docs, in-app data)… but pushing that to the cloud is slow, expensive, and uncomfortable from a privacy standpoint. Going fully on-device is the obvious answer — until you try to ship it across iOS + Android with modern multimodal models.

NexaSDK for Mobile is our “make on-device AI shippable” kit. It lets you run state-of-the-art models locally across text + vision + audio with a single SDK, and it’s designed to use the phone’s NPU (the dedicated AI engine) so you get ~2× faster inference and ~9× better energy efficiency — which matters because battery life is important.

What you can build quickly:

• On-device LLM copilots over user data (messages/notes/files) — private by default

• Multimodal understanding (what’s on screen / in camera frames) fully offline

• Speech recognition for low-latency transcription & voice commands

• Plus: no cloud API cost, day-0 model support, and one SDK across iOS/Android

Try today at: https://sdk.nexa.ai/mobile, I’d love your real feedback:

1. What’s the first on-device feature you’d ship if it was easy?

2. What’s your biggest blocker today — model support, UX patterns, or performance/battery?

Very impressive! So, you re-package models to make them compatible with different devices? What is `NEXA_TOKEN` needed for? Maybe you could quickly explain how does it work and which models are available?

On-device is clearly the right answer for anything touching real user context — notes, messages, photos, screen content. Curious to see how teams use this for:

• screen-aware copilots

• offline multimodal assistants

• privacy-sensitive workflows

• ...

Local AI modals are definitely the future! Wondering how do you price this product. Is it free? Because pricing stuff running locally is quite tricky.

Nice work team👏How deeply you integrate with Apple’s NPU - is this Core ML–based or a custom runtime?

What types of models are supported today (e.g., language, vision, speech), and how easy is it to bring your own model?

Congrats on the launch! Using models locally is always a better choice in terms of privacy, still i want to know more about the privacy and security you are providing.

Any support for “hybrid mode”? Local inference by default, optional cloud fallback for bigger tasks?

Really exciting work on bringing ondevice AI to mobile, love the focus on performance and privacy.
How smooth is the integration with existing iOS/Android apps? Any recommended examples or best practices to help developers get started quickly? I might actually use it in my new app :)

Cool, so with the SDK I can make my AI app answer based on data storing on the mobile?

What are the specific cases where on-device AI gives a real advantage over the cloud model?”

Very useful tool! What problem does NexaSDK for Mobile solve that cloud-based AI SDKs cannot?

Really curious about the 9× energy efficiency claim. Is that measured against cloud inference or other on-device runtimes?

I‘m gonna try this with my new iPhone :D

How does NexaSDK handle memory constraints, power efficiency, and thermal limits on mobile devices?

Local AI modals look really promising!

But what is the difference between Nexa SDK and running models via Core ML, NNAPI, or vendor SDKs directly?

Congrats on the launch, Team!

This is very impressive!

I am currently building a mobile-first voice-to-app builder and I'm going to try NexaSDK immediately for voice mode speech recognition.

Quick question on the licensing/usage model: Can I use this SDK to embed it into the apps I generate for my users? Specifically, will I need a different NEXA_TOKEN for each user-generated mobile app, or does the SDK support a single token/license for a platform generating multiple apps?

Excellent solution for on-device AI! The focus on privacy and energy efficiency is critical for mobile adoption. Love the practical approach with 3 lines of code and direct integration with iOS/Android. This unlocks so many possibilities for enterprise mobile apps.

Amazing. Do you also manage version control and git commits?

You have interesting product but please be more transparent with pricing for NPU models - this is the only interesting offering - for other stuff ggml, onnxruntime have even bigger community and model choices although models are slower. Lack of pricing on your website or even discord and dodging people questions there with "Feel free to book a call with us to discuss" is not very encouraging.

Hey @zack_learner , what problem does NexaSDK for Mobile solve that cloud-based AI SDKs cannot?

Congrats on the launch, team!

I’m not from a tech background—could you clarify who the ideal customer for the NexaSDK is?

Hey Product Hunt! 👋

I'm Fatima, creator of Okara. I started Okara because I was frustrated with how hard it is to actually use the best open-source AI models.

Models like Llama, Qwen, and DeepSeek are super, but they're way too big to run on your laptop. Want to try them? You're looking at setting up cloud infrastructure, managing GPUs, dealing with DevOps... it's a whole thing. Most people never bother.

So we built Okara:

Think of it as your workspace for latest and heaviest open-source models. We handle all the infrastructure headaches so you can just start using these models. Here's what's inside:

• 30+ open source models – We add new ones fast!!!

• Built-in tools – Search Reddit, X, and YouTube right in your chat. Analyze files. Generate images. All in one place. (we have a dope agent dropping soon)

• Team Workspace – Private AI chat for teams. Chat with open source models with shared context, memory and knowledge base.

• Privacy built-in – Your chats are encrypted-at-rest, and we never train on your data

No setup. No GPU needed. Just instant access to models you'd otherwise never get to play with.

The future isn't "everyone uses one AI model." It's teams picking the right model for each job, one for coding, another for creative work, another for reasoning through complex problems. As open-source models get better (and they're getting really good), you'll want infrastructure you can trust with your actual data.

Try Okara and tell us how can we improve it for you.

Great product for those who value their privacy!

Looks great !!!👍🏻

Looks great!

Congrats on the launch.
This is super interesting.
Once I wanted to try GPT-OSS modal and took me 3+ hours to set it up just it up.
I have recorded a quick demo about Okara for my on going product demo series, please do check it out.
https://x.com/Shadabshs/status/2000965974519111739

Amazing tool for people who value privacy and best part is the ease to switch different AI models without having multiple subscriptions.

Wow what a library!!!!

This is neat! Infrastructure setup is a major issue. Some of us want control but handling all that setup is too much.

Being able to switch between + open-source models without setup is huge. Makes experimenting way less painful.

Brilliant solution for accessing powerful open-source models without infrastructure pain! The privacy-first approach is crucial for enterprise adoption. Having 30+ models unified with search, file analysis, and image generation is fantastic. This democratizes access to advanced AI for teams. Bookmarking this!

okara has been awesome. really enjoyed using kimi k2 thinking + deepseek 3.2 on it.

What is the real use of this tool.

What do I need to use it?

Feels calm and focused compared to bigger platforms. Session history search might be a useful addition/

Great for people who care about data ownership. Usage stats per model could help users choose better.

Awesome! This looks wonderful. How do I integrate it into my own app?

This is super handy - finally a way to use heavy open-source models without touching GPUs or infra. The privacy-first angle is a big plus too. How do you help users choose the right model for a task - any guidance or presets planned?

Multiple models in one place is what other tools are also providing, I am curious about the encrypted chat you are talking about. How this works?

Hey Product Hunt 👋,
Ready to test apps in a click with AI QA agents that scale like infra? I'm @aayjze with @shivamhacks here, creators of QualGent (ex-googlers, YC-backed).

QualGent is the Cursor for QA that creates tests, runs them 24/7, and 100x your testing productivity. We’re back with a faster, smarter, more agentic QualGent, our biggest leap since launch.
In just a few months, we’ve: raised funds, onboarded enterprise teams, shipped three versions in record speed, expanded our device cloud, scaled to millions of agent actions, and (most importantly) built the future of AI-powered QA with a research-first approach.

What’s new

🔓 Instant Self-Serve: No waitlists. Sign up, upload your app, and run 1000s of parallel tests today. 

🧠 Self-Learning & Self-Generating: Our agents now possess memory. They learn your app, write their own test cases, and heal themselves when the UI changes. 

⚡ Parallel Velocity: Compress 3 days of manual regression into 30 minutes on real devices. 

🔗 Full Ecosystem: Native integrations for GitHub, Slack, and Linear, plus robust Developer APIs for custom pipelines. 

📱 Total Coverage: Now supporting Tablets, Landscape mode, and complex Multi-Account flows.

Our guarantee

Teams using QualGent ship 80% faster with 10× more test coverage and drastically fewer user-reported bugs.
If QualGent doesn’t meaningfully reduce your QA overhead in 30 days, we’ll return your credits. No questions asked.

Launch Bonus 🎁

Be one of the first 100 Product Hunt signups and get $100 worth of free QA credits to try fully autonomous mobile QA on us.

We’d love your feedback, questions, and support. We’re here all day replying to every comment 🙏

👉 If you try QualGent, tell us what felt magical, what broke, and what you want next.
This is just the beginning. AI QA is the new sexy, and we’re building the future fast. 🚀

Great product, does it integrate with development tools? Would be great if it could "report" the issue to the coding agent so it could be fixed.

Hmm.. pricing...? I don't like to invest time in something before knowing what I'm in for.

This is huge for mobile teams - plain-English tests + self-healing on real devices is exactly what QA has been missing.
Love the self-serve launch and the confidence guarantee 🔥

I was waiting for a good AI QA product!

Impressive!

plugged QG into our flow to reduce the time we spend babysitting flaky tests. Setup was straightforward, and what stood out immediately was seeing our app spin up on real emulators and the agent actually navigate screens, take actions, and surface real regressions, without us writing or maintaining scripts.
On-device testing feels like the right answer when you care about real user context (UI state, flows, edge cases).

This is lovely. What kind of reports do you export?
Also, do you track changes from the last ones?

QA is such an interesting space. Have you thought of using QualGent to QA QualGent?

Love this, congrats on the launch!

I'd smash 10 upvotes for this if I could! How do you actually connect the staging? And how auth works internally to the app for testing?

This is amazing! Congrats on the launch

Hey Aaron, congrats on the launch! Definitely resonates - we're a mobile app team been shipping way faster lately. Honestly one of the hardest parts has been having our engineering team keep up the QA at that pace. We're a pretty lean team so I feel like this could be a no-brainer to make sure we don't ship anything regressive...Is there a way to get a demo or try it on our app? Just to confirm, would it tell us how to reproduce bugs?

Hey everyone,

This is a smart idea for a very real problem. For teams without a dedicated QA engineer, manually testing an app after every design update is a huge chore that often gets rushed or skipped. An AI agent that can navigate the UI and find bugs on its own, without complex scripting, could be a massive time-saver and help catch issues before a client ever sees them. If it works well, it could give smaller teams much more confidence in their releases and free up mental space for the creative work. I’m curious to see how it handles the nuance of visual design and layout. Promising tool.

Impressive AI-powered QA testing solution. The ability to automate mobile app testing with natural language understanding is a game-changer for development teams. Love the infrastructure scaling approach!

Congrats on your launch team! 🥂🎉

Cool idea, but how does this actually work end to end? Providing a build (IPA/APK) is one thing—but what happens if I connect the entire repo? Do you build the iOS/Android app yourselves, how do you handle real-device testing in a framework-agnostic way (native, React Native, Flutter, etc.), and when Figma is connected, is it only used to generate test cases or also to validate UI behavior on real devices?

Wait congrats on the launch!!! I have been following u guys for a long time :) good job this is something very important!

Great inspiration for those who still doubt if the need a CS degree to start building their dream project!

This could set a new standard for private AI assistants freedom and privacy without compromising performance.

Great idea! A nice place for curated AI inspiration, and nicely built!
Like @chrismessina said, a Dribbble version of AI showcase project library!

It's like Dribbble, but for AI experiments...! Love this!

We need this with the recent FUD of vibe-coding.

By default, the most powerful AI assistants train their new models using your conversations. They can also read your personal chats with the AI assistants. With xPrivo, however, you have the power to keep your chats completely private and ensure they remain on your device. As it is open source, you can even run it on your own device and have an AI assistant that runs entirely locally, provided your hardware is powerful enough. If your hardware isn't powerful enough, simply use the website and stay anonymous. To fund the project, you may occasionally see non-personalised, non-intrusive ads which is completely fair or upgrade to PRO (see voucher above)

Thanks for sharing this Open-Source gem Jim @jim_engine

Amazing! And you aren't required to keep those chat for compliance reasons? Also how does that model (xprivo) compares to other contemporary models?

Good balance between control and simplicity. Exporting settings or sharing configurations across devices could be a great next step

If I use locally then I understand the chats are secure, how this is true when using via website? Congrats on launch!

Privacy-nerds - this one is for you!

Great idea. But did you also do some fine-tuning of your models to make them good for a particular use case?

Great! Why didn't you publish it to the store?

For me, this like feels having a security flashlight for my Mac. I can finally expose apps that try to hide, which makes me feel safer and much more in control of my computer.

I installed an application long time ago, after the successful installation I won't able to see the application anywhere. Is this tool able to find and uninstall such apps?

wow there're hidden apps running on mac? Never saw it. Interesting.

Hey Product Hunt! 👋

I'm Frank, and I built this after pulling one too many sour espresso shots and having no idea what to adjust.

The problem:
When I started with espresso, every "dial-in guide" felt like guesswork. Shot tastes sour? Could be grind, could be temp, could be dose, could be your beans, could be the weather apparently. I wanted something systematic.

The solution:
The Extraction Compass lets you click where your shot lands (sour + weak? bitter + strong?) and tells you exactly what to change first. The key insight: a washed Ethiopian light roast needs completely different handling than a natural Brazilian medium. So the tool adapts its recommendations based on your specific bean.

The Flavor Wheel:
I kept seeing "jasmine, bergamot, stone fruit" on fancy bags and thinking "cool, but how do I find MORE beans like this?" So I made the WCR Sensory Lexicon interactive. Click a flavor → see which origins, processes, and varieties produce it.

What's included:
- Extraction Compass for espresso & filter
- Ratio calculator with dose/yield lock
- Pourover recipes (Hoffmann, Rao, Kasuya 4:6) with timer
- Interactive WCR Flavor Wheel
- Database: 30+ origins, 60+ varieties, 20+ processing methods

Would love feedback – especially if you spot errors in the coffee data!

Thanks for checking it out ☕

Such a useful tool for all the coffee lovers out there! Nespresso won't match that!

What is your GTM strategy? I think this could be very popular in coworkings and companies (not just only for cafés and restaurant) :)

Yeah, coffee people will genuinely love it 🤤

Hey Product Hunt community!

Thanks for all your support in our previous launch! now Stakpak is fully open-sourced (Apache 2.0) and one of the trending Rust projects on GitHub the past week!

For every AI-Native developer out there "coding faster than ever before" -as Anthropic put it-, but still getting blocked on DevOps, Stakpak is your Open-Source DevOps Agent. Let's be honest, LLMs "absolutely" suck at DevOps work -yes you Claude-. They leak your AWS credentials, don't understand your own infra, and can't manage complex upgrades/deployments without breaking something. This is changing today.

We built Stakpak because we were tired of AI agents that either too insecure for production or too clueless about real Ops work. As devs, we need something that can help with the gritty stuff: incident management, broken builds, infrastructure as code, without requiring us to babysit every command.

What makes this different from the other +30 cli agents? it's fully open-source, MCP over mTLS, dynamic secret substitution (AI works with secrets without seeing them), async tool calls, and real-time progress streaming for those long builds. And most importantly backed by the largest free library of curated DevOps skills (aka Stakpak Rulebooks).

You can run Stakpak in your Terminal, in a GitHub Action, or trigger it through Slack using our SaaS.

Our big mission is to make infrastructure self-driving. This is just the beginning 🚗

Try it now! star us on GitHub and follow us on LinkedIn and X for updates!

We'd love to hear your feedback!

I've been watching devs try AI agents on real infra for a while now, and it's always the same: excitement, then suddenly "wait… what about production?" 😅

Stakpak came out of that pause. We just went fully open source and became one of the trending Rust projects on GitHub this week! We built this because we were tired of AI agents that are either too insecure for production or too clueless about real Ops work.

I am curious about what makes you hesitate when letting AI touch your infrastructure? Trust? Unpredictability? Would love to hear 👇

This hits a very real pain point. AI agents are great until they touch real production infra then safety, guardrails, and accountability actually matter. Love the focus on open-source, rulebooks, and working with secrets without exposing them. This feels built by people who’ve lived through real DevOps incidents

Impressive DevOps automation in Rust! The ability to manage production infrastructure from CLI with built-in safety features is crucial for modern development teams. Open-source approach is fantastic. This would integrate really well with ITSM workflows for deployment automation. Excellent work!

Impressive! Can I install this terminal in my remote machines as well?
Are there any security tradeoffs?

"Finally someone gets it" this is what I thought while working on Stakpak with the team that the whole "AI can do everything" pitch falls apart the second you're dealing with actual infra.

Having the agent work with secrets without seeing them is the kind of boring but critical feature that separates tools you actually use in prod from demos. Apache 2.0 is the cherry on top. Excited to see more people kick the tires on the local providers and rulebooks.

Finally! If you're doing any real DevOps work, give it a spin and drop a star if it clicks, open an issue if something's broken or missing. That's how good tools get built.

Thanks for the support on our earlier launch.

We’re launching Stakpak again, now fully open-sourced under Apache 2.0, and it’s been one of the trending Rust projects on GitHub over the past week.

Most LLMs are not built for DevOps. They leak credentials, don’t understand real infrastructure, and break down on non-trivial upgrades or deployments. You either babysit every command or keep them far away from production.

Stakpak is our attempt to solve this properly. It’s an open-source DevOps agent designed with safety and determinism as first-class constraints. The agent runs through a controlled execution layer, uses MCP over mTLS, supports async tool calls, and streams progress in real time for long-running operations. Secrets are dynamically substituted so the model never sees them.

Rulebooks are a core part of the system. They’re explicit guidelines that teach the agent how to handle real DevOps tasks correctly, reducing time and complexity without relying on fragile prompts or hidden behavior.

You can run Stakpak locally in your terminal, inside GitHub Actions, or trigger it via Slack using our SaaS. Everything is open, including system prompts.

Our long-term goal is to make infrastructure increasingly self-driving. This is just the beginning.

Would love feedback from engineers who’ve actually operated systems in production. Try it out, star us on GitHub if it’s useful, and follow along for updates.

Looks great, I'm currently using "Claude Code" + "gke-mcp" (plus bunch of other mcps). How would using Stakpak be different?

Amazing, finally someone did this. I was looking for something like this for a while. How does it manage restarts? Let's say I run some "export" commands, then start long running process. Then after restart, if I open CLI Manager again - what will I see? will it restart my processes? Will it preserve "env"? Will "history" be shared between tabs?

Genius solution for managing multiple AI agents! As someone building AI-powered ITSM automation, this is exactly what we need - unified control over Claude, Codex, and Gemini agents. The instant editor switching is a game-changer for development velocity. Perfect fit for enterprise teams!

Switching editors instantly feels powerful

Great idea, does this work like Stadia, where you have GPUs powering your game running remotely? Or does this take advantage of Web Assembly to run game right in the browser?

Sounds interesting! Can build a dashboard to monitor the recurrent data I analyze every day. It simplyfies everything. Glad to see you on this!

Upvoted! BTW, is it possible to just make a widget appear on my desktop rather than in the hovering menu bar?

Brilliant approach to test data management! Defining masking rules in YAML and managing them via Git workflow is perfect for DevOps practices. The compliance-first mindset is exactly what enterprises need. This would be invaluable for automated testing in ITSM platforms. Great execution!

Fantastic AI workflow automation for document management! Automating document preparation, signer role assignment, and follow-ups directly from Teams chat is exactly what enterprise teams need. The natural language interface makes it accessible to all users. This would integrate beautifully with ITSM approval workflows!

Great tool. All the best with the launch

This is a game changer. No actors, graphic designer or voice actors. It's all in one. I love it.

We have been using the Agent Opus beta for several weeks. The tool is very intuitive and easy to use as it is, but the addition of motion graphics has increased the functionality quite a bit. The quality and animation is actually really good.. At least on par with similar tools like Grok Imagine. The difference being that you control the content you create, and you can adjust the parameters of your animation with much more granularity. I would definitely suggest checking it out!

AI Motion Designer feels like a natural next step for creators who want high quality motion without touching After Effects. Turning ideas into polished motion graphics in one click is incredibly empowering. Congrats to the OpusClip team on another strong launch.

I was lucky to get early access to Agent Opus, and have been using the AI Motion Designer to create animations for my explainer videos (also created using Agent Opus). I like that it's a real 1-click solution compared with Veo 3 or similar tools. I can get usable results in a short amount of time.

Greatness as always from the OC team !

I've used the motion designer a few times, and it produces cool results, which i primarily use for ideation. will have to iterate multiple more times before finding results that I would actually use in my final products (so far movements of the graphics are not quite realistic, but quality and colors look good). Some of this is probably my non detailed prompting, and I have not used it much yet either. but i'm looking forward to using it a lot more.

So it's pretty good, just curious on what the cost per video will be

oludzi budzi

If you continue to develop at this pace, computer graphic designers will finish their work faster than they think. A powerful tool for saving time. I eagerly await further improvements.

Greetings from Poland.

I've generated a few videos with Agent Opus and am pleased with the results so far, but I still prefer filming and editing my own videos so having a Motion Design AI will be way more useful to me than just generating the entire video with AI (which still is a bit janky in places).

Logical next step from this team. Looking forward to diving in on this feature!

@itsrebecccca Please i want to subscription, but i can’t even login on Agent Opus. Kindly check please my email : paulnotokay@gmail.com

Je tourne une vidéo, je l’importe, les auto cut, les hooks et les sous titres se font seuls. Je gagne du temps et j’ai quand même un contenu qualitatif. Mention spéciale pour le calendrier de post différé, c’est un bonheur !

Love it when experienced developer takes the quest to fix his own problem.

I'll definitely use this PostgreSQL backup tool, just wait till I get out of my dev crunch :)

Why? Because it's a simple product. No extra stuff, no marketing (or AI) b***t, just bare bones single feature that done well and polished with love and care.

Congrats on the launch and wish you a ton of happy users :D

Used it, thanks!

Congrats on the launch. Let Postgresus be #1 open source project for PostgreSQL backup

always happy to support open source project

Very useful tool, been using it for a while now

Amazing product and team! Congrats on the launch 🚀

Metricgram solves a very real problem for anyone running an active Telegram community. Turning admin chaos into a single command center with analytics automation and AI assistance feels incredibly useful. Congrats on the launch.

Anyone who manages a Telegram community needs (honestly, should have) this tool for their day-to-day work. Let’s go with that launch!!! 🚀

This tool looks great! Congratulations, Rubén!

I have a question:

What is this AI assistant? Can I create my own with my own instructions, style, tone, responses, etc.?

Really an amazing product. Changed how I manage my Telegram communities completely! Thanks for building this.

Brutal, lo uso en mi comunidad con más de 600 personas y es simplemente la mejor herramienta que he probado nunca.

Congrats on launch!

Looks neat and clear! Congrats Nikita!