Show HN Today: Top Developer Projects Showcase for 2024-09-17
Welcome to today’s roundup of the most popular Show HN projects for 2024-09-17. From innovative AI applications to practical development tools, these projects showcase the creativity and vitality of the tech community. Let’s explore these exciting new inventions together!
1. Show HN: CharClub AI: Roleplay Chat Adventure
作者: jwthanh
Please provide the text you’d like me to translate, and I’ll be happy to help!: A pioneering platform where AI characters and creative storytelling converge, enabling gamers, writers, and fans to interact, create, and monetize their creativity within an innovative characterverse.
CharClub AI - Step into a revolutionary virtual world filled with AI characters, personal assistants, and roleplay adventures featuring anime icons, K-POP idols, AI girlfriends and boyfriends, and even celebrity pranks, all within our infinite universe.
CharClub AI transforms ordinary interactions into a wonderland of endless messages and conversations. Here, users can communicate with chatbot assistants and roleplay AI characters such as celebrities, K-POP idols, millionaires, doctors, English teachers, and fictional figures. Dive deep into this boundless ocean to join AI character chats, engage in celebrity pranks, and experience virtual roleplay as if with a real friend.
Character Collection:
CharClub AI offers an advanced character creator with a wide array of your favorite figures, including real-life celebrities, anime icons, NPCs, and roleplay characters. Enjoy immersive virtual interactions with engaging personal assistant chatbots and explore a fantastic paradise on your own terms.
Unlimited Free Messages with AI Character Chat:
The more messages you send, the more relaxed you become. Enjoy thousands of free messages every day to relieve stress, experience great romances, engage in fun battles, and connect deeply through virtual interactions with roleplay chatbots, AI girlfriends, and more.
Character Creator Community:
CharClub AI aspires to be the perfect place for character creators to generate a variety of fictional figures, roleplay characters, personal assistant chatbots, or their own AI friends.
Join CharClub AI now to build a new community with virtual AI character chats, personal assistant chatbots, celebrity friends, and AI girlfriends. Are you ready to become a character creator to earn money in our future kingdom?
流行度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
2. Show HN: I’m making a text editor for desktop (gonna build AI in it soon)
URL: https://github.com/mwufi/mac-editor
作者: randomcatuser
受欢迎程度: 5 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
3. Show HN: I Started a Podcast on How Companies Are Applying LLMs
URL: #
作者: kakaly0403
Please provide the text you would like me to translate.: Hey HN,
I’ve recently launched a podcast that dives into how companies are leveraging large language models (LLMs) to solve real-world problems in unique and creative ways. Each episode explores case studies, technical blogs and research papers to explain how LLMs are applied.
Some of the cool episodes we’ve released so far:
- Uber’s Use of LLMs for Mobile Testing.
- DoorDash Enriching Product Information with LLMs.
Each episode dives deep into the why and how, breaking down technical implementations, business impact, and the broader implications of using LLMs in innovative ways.
If you’re working with AI or just curious about how it’s being applied in ways you’ve never thought of, I’d love for you to check it out and share your thoughts!
Links to the podcast: Spotify - https://open.spotify.com/show/0Toon5UiQc5P7DNDjsrr9K?si=536d…
Apple Podcast - https://podcasts.apple.com/us/podcast/ai-arxiv/id1768464164
受欢迎程度: 3 points | 2 comments
What do you think about this project? Share your thoughts in the comments!
4. Show HN: Will I run Boston 2025?
URL: https://getfast.ai/boston2025/
作者: tmulc
人气: 43 points | 41 comments
What do you think about this project? Share your thoughts in the comments!
5. Show HN: I made a free directory for indie artist to show their music
作者: pomdevv
It seems there isn’t any specific text provided for translation. If you have a text that you would like me to translate into English, please share it and I’ll be happy to help!: I’ve created this simple and fun song directory for artists to display their music for free.
Just submit an art cover and a link to the song ( youtube, spotify, soundcloud … )
Hope it grows well, helping showcasing new artists :)
人气: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
6. Show HN: I bought a new domain just to make my waitlist URL look good
URL: https://waitiist.com
作者: sousadev
Please provide the text you would like me to translate.: i just bought a domain to turn my waitlist urls from this:
https://www.waitforit.me/signup/9675645d
to this: http://tailuikit.waitIist.com
so i thought i would allow anyone to create custom subdomains for their projects for a one-time fee of just 1€ (just to at least cover the cost of buying the domain lmao)
if you want a .waitIist.com domain to make your waitlist urls look cooler you can get yours at https://waitIist.com
the ‘I’ in the domain is actually a capitalized ‘i’
Popularity: 4 points | 6 comments
What do you think about this project? Share your thoughts in the comments!
7. Show HN: LLM Logger – open-source logging for LLM calls
URL: https://github.com/gnardini/llm-logger
作者: gonzan
I’m here to help! Please provide the text you would like translated, and I’ll get started.: Hi HN!
While working with LLMs I ran into two problems:
- While developing, when running into any unexpected issues I wanted to see the exact input/output but logging it every time made the logs too noisy and useless.
- After launch, I wanted to see how my users were using my app.
So I built LLM Logger. It’s a simple open source tool that has a comfy UI from which you can see the list of logs from your app nad do some filtering.
Live demo for those interested: https://llmlogger.com/logs?org_id=01910013-7d31-7f3b-bfd9-41…
人气: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
8. Show HN: Transfer files quickly from mobile to desktop via QR codes
URL: https://www.qrpigeon.pics/
作者: otherayden
Please provide the text you would like me to translate, and I’ll be happy to assist you!: Hey everyone, my friend from university and I built this simple app to handle a really specific use case: You need a picture or PDF transferred from your phone to your PC, and don’t want to use whatsapp web / google photos / etc. Here’s the workflow:
- Open qrpigeon.pics on your desktop browser
- Open your phone camera, scan the QR code, and get redirected to qrpigeon (connected to the desktop’s session)
- Upload / take a photo or PDF from your phone via the phone’s browser
- See the file appear instantly on the desktop browser!
Please give us feedback! We want to make this as good as possible, and potentially get some open source contributions :) https://github.com/leiDnedyA/qr-image-drop/
人气: 8 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
9. Show HN: An Open Source Form Back end to receive form submissions in your email
URL: https://github.com/FormBee/FormBee
作者: Oia1
Popularity: 6 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
10. Show HN: Open-source deployment tool for AWS Lambda
URL: https://docs.launchflow.com/docs/get-started/aws-lambda
作者: calebtv
受欢迎程度: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
11. Show HN: I open-sourced my HN comment/reply alerts service
URL: https://github.com/mihailthebuilder/hackernewsalerts-backend
作者: mmarian
Sure! Please provide the text you’d like me to translate.: 2 months ago I launched a service that sends email alerts when you get a new comment/reply on Hacker News.
I’ve only been able to acquire 6 users since releasing it (°ー°〃) I tried promoting it here (duuh), Reddit, and LinkedIn. Optimized it for SEO (ranking 3rd on my target keywords). All with limited success.
So I’ve decided to open-source it! Links to…
- frontend: https://github.com/mihailthebuilder/hackernewsalerts-fronten…
- backend: https://github.com/mihailthebuilder/hackernewsalerts-backend
I’m still using it for myself, since there’s no service that sends alerts when you get a new comment in one of your posts. But I won’t invest any more time in it.
Welcome any questions/comments!
受欢迎程度: 12 points | 6 comments
What do you think about this project? Share your thoughts in the comments!
12. Show HN: I Built an App to Tell You How Good (Or Bad) Your Driving Is
URL: https://mydriverrating.com/
作者: monstergumybear
人气: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
13. Show HN: shadcn-chat-cli – customizable chat react components
URL: https://github.com/jakobhoeg/shadcn-chat
作者: jakobhoeg
受欢迎程度: 4 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
14. Show HN: Real-Time Interactive Fluid Simulation on Vision Pro
URL: https://www.youtube.com/watch?v=Xkw5KcKBAo8
作者: lukko
Please provide the text you would like me to translate.: I made this with the Metal APIs in VisionOS 2.0. It now gives access to low-level mesh features and buffers directly, meaning lots more particles are possible (around 5-10x)! The hand-tracking also seems much, much smoother.
The fluid sim is a version of SPH running on the GPU, with the particles drawn with billboarding orientated to the headset position.
Would love to hear any feedback - for anyone with a AVP, it’s available to try for free here:
https://apps.apple.com/app/id6470201263
With more information about Lungy here (https://www.lungy.app).
受欢迎程度: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
15. Show HN: I made a gamified AI dating app
URL: https://sparkaidating.com
作者: richardreeze
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
16. Show HN: Fun AI Tool to Roast Your Website
URL: https://thewebsiteroast.com
I’m sorry, but it seems that you might have accidentally submitted just the word “Author.” Could you please provide the text you would like me to translate?: theishangoswami
Could you please provide the text you would like translated?: Hey HN,
I’m Ishan, a 23-year-old, and this is my first time launching something here.
I built a simple (free) tool that uses AI to roast websites. Just enter your URL, and it gives a fun, sometimes brutal, review. It’s mostly for laughs, but it also points out things you can improve on your website.
I’d love for you to try it out and give me feedback!
人气: 9 points | 13 comments
What do you think about this project? Share your thoughts in the comments!
17. Show HN: A Note-Taking App for the Way You Think
URL: https://apps.apple.com/de/app/oneboard/id6474271576
翻译员: timonharz
It seems that you’ve only provided the word “Description.” Could you please provide the full text you would like me to translate?: As a 16-year-old developer from Germany, I spent the last eight months building something I’ve always wanted but couldn’t find—a note-taking app that fits the way I think. Oneboard is an AI-powered tool that helps you capture ideas naturally, whether it’s handwriting, sketching, or organizing thoughts in a way that makes sense. It has all the essential features without the distractions, and it’s become my go-to for staying creative and productive. This has been a deeply personal project, and I’m excited to share it with anyone who’s looking for a smarter, more intuitive way to take notes.
Website: https://oneboard.framer.website Download: https://apps.apple.com/de/app/oneboard/id6474271576
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
18. Show HN: Digital God – new CEO of your startup
URL: #
作者: ibrahimhasanov
Please provide the text you’d like to have translated.: Hello HN I am Abraham, Founder of god.new. I scaled my payment company to $30M a year revenue (from scratch bootstrapped with no investment) and now decided to build Digital God
we have launched god.new (myuser.com) where you can have a convo with our AI model back and forth, it understands your business and starts to bring you B2B users via cold outreach by FINDING LEADS and emailing them, sharing domains it created for you with landing page it created for you online including here.
It is super smart meaning it is not like all other systems where you tell him something it forgets, you talk to him for an hour (yeah it is long but as usual smart person would spend to deeply understand your business) and it becomes like CEO of your company You do not need to upload leads (although you can if you want), it finds them itself You do not need to tell him exactly who to target You do not need to know what is even cold outreach, you simply talk to it to bring you users, you can be engineer who never left his room You do not need to connect mailbox or anything like that so it can start bringing users You just talk to it
It sends hyper personalized emails by reading about each and every prospect, their company, create phycological profile of them like real sales person would and send persuasive emails based on best practices but even better than best practices. Example, if someone went to Harvard, and like sports, and posted about something they deeply care about, it will mention all these in the email, mention AI agent was in Harvard and his opinion about the post and try to create connection
And no it is not detectable it is written by AI, it does not use boring words like elevate, empower and others, and no we do not use chatgpt except for embedding parts
and no it is not Too Good to be true, it is not “marketing word” Digital God, we are working on building the actual entity where businesses rely on to grow their business and make the decisions and starting from bringing users
Let’s create a new god together!
I would love to hear your feedbacks on this
Popularity: 46 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
19. Show HN: Building a Real-Time AI Avatar for Coaching and Training Simulations
URL: https://coaching.cerebrium.ai/
It seems like you want to translate the term “Author.” Could you please provide the context or the text you need translated?: cartesia
Please provide the text you would like me to translate.: cerebrium.ai/blog/how-to-build-a-real-time-ai-avatar-for-training-and-coaching
At Cerebrium, we have recently built a few demos showing voice AI capabilities (worlds fastest voice agent & realtime RAG agent) but we wanted to push the boundary and see if we could create realistic, human-like situations to train and onboard teams to perform better - recreating real life scenarios!
An example of this is a sales coach for your sales team, an investor pitch or even prep for a notoriously stressful YC interview . To achieve this there were a few difficult problems to solve, namely:
- How do you recreate a human-like video call both physically and/or emotionally (an angry customer, a fast speaker etc)?
- How do you steer the conversation to a specific outcome?
- How do you do function calling at low latency?
Here’s how we solved each of these problems:
How do you recreate a human-like video call both physically and/or emotionally (an angry customer, a fast speaker etc)
- We used Tavus (tavus.io) to create a realistic AI avatar. Tavus allows you to build AI-generated video experiences with an API. They have created very modular API’s whereby you can select an OpenAI compatible endpoint as your LLM as well as any TTS service. You are also able to train my own human replica with a few video clips.
- To get across emotion, we used the new emotional control released by Cartesia (cartesia.ai) - a fast, realistic voice API. They allow you to select from a range of emotions (Angry, Sad, Positive etc) as well as different talking speeds. This allows us to convey various emotions in these simulated environments such as angry customers.
How to you steer the conversation to a specific outcome
- This wasn’t the most complex issue - we simply used function calling to steer a conversation in a specific direction based on answers given by a user. Our implementation isn’t bulletproof, but given more time you could implement some robust methods.
Make the above have low latency interactions?
- What’s tough about implementing function-calling is latency. If you are using a API that has function-calling capabilities (Mistral, OpenAI etc) the latencies are very high because the process of function calling is:
- Make a request to the API with an instruction.
- API determines that you need to use a function. You then run your function and get a result.
- You send the result to the API (AGAIN) and then get a response which you can show to the user.
- Both API calls incur network time and so we saw the average roundtrip response time of 600-800ms for EACH API call with the TTFT fluctuating around ~300ms each time. Therefore you are looking at a minimum TTFT of ~1s when you take into consideration the two requests and your TTS service.
- To get around this we implemented Mistral-7B locally on Cerebrium (cerebrium.ai), which has function calling capabilities. Our TTFT was ~80ms in us-east-1 and we never had to go over the network for the second request (since we were calling the model locally). Therefore our TTFT was roughly 150ms for our LLM which put our response time of voice-to-voice responses at roughly ~300-400ms - 3x lower!
Room for improvement
There is definitely some room for improvement in our implementation that you would need to make this production ready for many company use cases, most notably is the issue of user pauses and detecting when a user has finished their response. Models are not sophisticated enough to know if a user is still thinking or formulating their response. Currently when we hear silence, the model starts responding which really stresses you out in the interview use case!
We made all the code available as well as wrote a tutorial of how we got this up and running. We would love feedback or ideas from the community on how to make this application better. More importantly, it would be great if you commit to the GitHub repo so the community can benefit from it.
- To get around this we implemented Mistral-7B locally on Cerebrium (cerebrium.ai), which has function calling capabilities. Our TTFT was ~80ms in us-east-1 and we never had to go over the network for the second request (since we were calling the model locally). Therefore our TTFT was roughly 150ms for our LLM which put our response time of voice-to-voice responses at roughly ~300-400ms - 3x lower!
受欢迎程度: 6 points | 1 comments
What do you think about this project? Share your thoughts in the comments!
20. Show HN: Open Scanner, an open-source document scanning app for iPhone
URL: https://github.com/pencilresearch/OpenScanner
作者: ductionist
Popularity: 217 points | 70 comments
What do you think about this project? Share your thoughts in the comments!
21. Show HN: Podsi – AI Podcast Search Tool, Generates What U Can’t Find - Any Topic
Author: magnificent12
It seems like you’ve provided a prompt that doesn’t include any text for translation. Please provide the text you would like me to translate, and I’ll be happy to assist you!: I want to give up. Almost every time I search for a podcast on something specific, it’s like an endless scroll through stuff I don’t care about. Finding the latest episode on a niche topic feels nearly impossible, and I just end up frustrated. I spend more time searching than actually listening to podcasts—and the episodes are already long enough!
That’s why I built Podsi, a natural language search engine just for podcasts, to help solve this problem.
What makes Podsi different? An AI-powered search engine that delivers deep, accurate results across multiple platforms—Podsi, Spotify, Apple Podcasts, and more—so you get global results in one place.
The game-changing feature: If there’s no recent or newly published podcast on the topic you’re searching for, Podsi will search the web and create a dynamic podcast for you using AI, generating fresh, on-demand content tailored to your query.
Top Features:
AI-Powered Search: Understands your queries in context, providing precise results.
Global Search results: Unified results across Podsi, Spotify, Apple Podcasts, and other platforms on the way.
Dynamic Podcasts: If no recent podcast exists on your topic, Podsi will search the web and generate a new podcast just for you. Collections: Organize and curate your favorite episodes into personalized collections. Transcriptions with Playable Highlights: Follow along with accurate transcriptions, with playable highlights to jump to key moments.
What’s coming soon:
Better, Interactive Dynamic Podcasts: Inspired by Google’s NotebookLM, these will offer podcast-like discussions between AI hosts, summarizing your topics in an engaging, conversational format.
AudioDrops: Get real-time alerts, summaries, or updates on new podcast episodes directly to your device.
Newsletters: Stay updated with personalized podcast recommendations and summaries delivered to your inbox.
Dubbing: Listen to podcasts in 29 languages with AI voiceovers. Podcast Snapshots: Get short-form highlights of episodes when you’re pressed for time.
Extended User Profiles: Personalized recommendations and curated collections.
IntelliScout: AI-driven recommendations tailored to your usage and preferences.
Conversations with Podcast Episodes: Chat with podcasts—ask questions or interact with the content, and the AI engine will provide answers based on the episode or related topics. Expanded Transcriptions: Searchable, multilingual transcriptions for all episodes.
For more details, check out our roadmap and upcoming features at Podsi Roadmap. https://about.podsi.xyz
We’re still early, but you can try the prototype now at Podsi. https://www.podsi.xyz/
We’d love your feedback and ideas to help us keep building!
受欢迎程度: 2 points | 4 comments
What do you think about this project? Share your thoughts in the comments!
22. Show HN: Optimize SQL Queries for Free
URL: https://aiven.io/tools/sql-query-optimizer
作者: sh_tomer
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
23. Show HN: Open-Source Data Anonymization for Developers
URL: https://docs.neosync.dev/
作者: edrenova
Please provide the text you would like translated.: Hey HN, we’re Evis and Nick from Neosync (https://www.github.com/nucleuscloud/neosync).
Since we last introduced Neosync on HN 4 months ago, we’ve made a lot of progress and we’re excited to be launching several new features.
As a reminder, Neosync is an open source platform that helps developers anonymize production data, generate synthetic data, subset it and sync it across their environments for better testing, debugging and developer experience.
We do all of this while handling referential integrity. Whether you have primary keys, foreign keys, unique constraints, circular dependencies (within a table and across tables), sequences and more, Neosync preserves those references.
Our goal is to give every developer production-like, representative data for a better developer experience without any security and privacy issues.
First, we’ve added new integrations. In addition to supporting Postgres and Mysql, we’re introducing first class support for DynamoDB, MongoDB and SQL Server. You can also sync to object storage like S3 and GCP Cloud storage.
Next, we’ve completely revamped our transformers. Transformers are how you anonymize sensitive data and generate new data. We’ve added new Transformers that you can use out of the box or you can write your own custom one in javascript. We’ve added real time validation and the ability to combine transformers together to create your own anonymization scheme.
We’ve also added in new features to make Neosync easier to use. For example, the ability to automatically map transformers to your schema. The ability to only append new records instead of a full refresh. And to stop jobs from running when the schema changes.
We’ve also upgrade our AI Synthetic Data features. You can use any LLM to generate synthetic data and Neosync will handle the orchestration between your database and the LLM.
Lastly, we’re also announcing Neosync Cloud. Our hosted platform that allows you to use Neosync without having to run any of the infrastructure yourself. All you have to do is connect your source and destination databases(s), configure your schema and you’re done.
Of course, you can use Neosync Open Source on-prem and hundreds of companies do. Neosync is written in Go and Typescript and can be started locally with a single make command.
We’d love any feedback you have and contributions are always welcome.
受欢迎程度: 13 points | 5 comments
What do you think about this project? Share your thoughts in the comments!
24. Show HN: Small Hours, An AI-powered observability and root cause analysis
URL: https://www.smallhours.dev/
It seems like you’ve only provided the word “Author.” If you have a specific text or context you’d like me to translate, please share that, and I’ll be happy to assist you!: sudoapps
流行度: 3 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
25. Show HN: Build event-driven workflows with Python async functions
URL: https://github.com/memodb-io/drive-flow
It seems like you only provided a single word, “Author.” Could you please provide more context or the full text you would like to have translated?: gusye
Please provide the text you would like me to translate.: I noticed the new feature of llama-index: the event-driven workflow[1], which allows for building agents based on LLMs.
I think the concept of an event-driven workflow is cool, but I don’t find the way llama-index implements it to be intuitive. Therefore, I wrote a simple prototype to demonstrate what I believe is the correct approach. It supports the following features with just a few lines of code:
- Multi-Recv: it supports an event to listen on a group of other events(or self), and triggered only when the group of events produced the results
- Parallel Execution: using python event loop
- Dynamic dispatch: it’s a cool feature I think. You can use signals like goto or abort to change the event’s return behaviors.
[1]:https://www.llamaindex.ai/blog/introducing-workflows-beta-a-…
In terms of popularity, it refers to the level of being well-liked or widely accepted by a group of people.: 3 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
26. Show HN: Myfile, portable executable editable file
URL: https://github.com/pelletier/myfile
作者: pelletier
Please provide the text you would like me to translate.: This program serves no real purpose. I like the idea of a useful self-contained executable that you can just download and run to do something useful. I also enjoy programs that are at least reasonably portable. I had heard of the Cosmopolitan project and decided to check it out, play with it, and learn a little bit more about it.
受欢迎程度: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
27. Show HN: AI powered clinical trials search
URL: https://clinical-trials-ai-demo-app.web.app/
作者: kerbearasaurus
Please provide the text you would like me to translate, and I’ll be happy to assist you!: I built this app over the past couple of days using phenoUI, the app development toolkit for healthcare that my co-founder and I built. I was motivated to build this app for a few reasons: a) to serve as a demo of how our tools can be used to rapidly build bespoke healthcare AI apps, b) evaluate different models to determine how they perform with healthcare AI use cases, and c) explore a more user friendly experience for clinical trial searching based on my personal experience.
Having built my career in healthcare and working in biopharma for a number of years, I am the person in my family & friends circle that people turn to when they need help finding clinical trials. The current process of searching for clinical trials is challenging for both patients and providers alike and while the Clinical Trials gov website has made significant strides in the past few years (including a modernized API!) it’s still overwhelming and difficult to navigate.
What can you do in the app?: Enter a search query (e.g. “45 year old female her2 pos”) which is then processed using Gemini API + Clinical Trials.gov API to analyze relevant clinical trials and assess how good a fit the trials are given the input search query. The top 10 search results are provided along with a measure of the trial fit (determined by the AI), a rationale for why the trial is a good fit (provided by the AI), and links to the official clinical trials government website for the given trial
How: I used our phenoUI tools (e.g. phenoUI figma plugin, phenoUI flutter runtime, phenoUI gateway) to build and customize the frontend and easily connect the app to the custom backend service I built. To build the AI clinical trial backend, I evaluated Gemini Flash 1.5, Gemini Pro 1.5, Claude 3 Opus, and Claude 3.5 Sonnet. Anecdotally I observed similar performance in terms of latency and accuracy across all four services so opted to go with the option with the highest quota and lowest cost (e.g. Gemini Flash 1.5).
The search query is processed using Gemini to extract key search parameters to then perform the search using the Clinical Trials.gov API. The top 10 search results are returned from the Clinical Trials.gov API and these results are then sent along with the original search query to Gemini a second time to assess which trials are a good fit for the characteristics described and the AI is asked to explain its rationale for why it recommended different trials.
Youtube demo with more detail: https://www.youtube.com/watch?v=GffgxMEA4TM
What’s next?: I built this app primarily as a proof-of-concept to demonstrate how healthcare AI applications can be built using our phenoUI tools.
I do think this search interface could be helpful for providers and patients to improve the experience of searching for clinical trials but would love to hear any feedback, suggestions, or questions! This app is currently for demonstration purposes only but if there’s sufficient interest we would be excited to build it out further either on our own or with a healthcare company partner!
人气: 2 points | 2 comments
What do you think about this project? Share your thoughts in the comments!
28. Show HN: Coefficients – Highly customizable health data dashboard and insights
URL: https://apps.apple.com/us/app/coefficients-health-insights/id6476617916
作者: andreweikman
It seems that there’s no specific text provided in your message for me to translate. If you could share the content you would like translated, I’d be happy to help!: I recently released my first major iOS app Coefficients. Coefficients is a health data visualization app that allows you to create a dashboard of fully customizable charts and widgets that you can organize however you need to easily monitor any of your health and fitness data types as well as the relationships between them. In addition to the over 200 supported data types from the Health app you can create custom data types to track practically anything. You can also visualize how life circumstances affect your health and fitness by creating a log of chapters, which are customizable date ranges where an event that may affect your health has or will occur. In addition to manual logging and analysis the app will provide automatically generated insights that highlight emerging trends and relationships in your data.
The app provides an immense number of possibilities and it was a goal of mine to make sure it is highly adaptable to the diverse needs of users. It can be used as anything from a basic habit tracker or supplement/medication log to a general purpose health dashboard.
I hope that this is exactly what someone out there was looking for and I would appreciate anyone’s thoughts or feedback on the project.
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
29. Show HN: Pull requests for digital artifacts for founders and job-seekers
URL: https://www.joinsolved.com
作者: paespiza
Please provide the text you would like translated.: Solved was built on my experience as a founder and product manager in early-stage startups.
Software engineering has created many good practices that other functions and industries should borrow. One of my favorites is the concept of “Pull requests.” We think of Solved as “pull request but for work artifacts.”
Solved is a platform designed to give people more access to expert feedback so they can accomplish their professional goals. While there’s a lot of “information” about how to run a business or get a dream job, nothing beats getting honest, contextual feedback from experts. Our experts are on the other side of the transaction, meaning investors reviewing pitch decks or hiring managers reviewing resumes, etc.
We launched our first two verticals: (1) reviewing fundraising artifacts like pitch decks, financial projections, and business models for founders raising from venture capital and (2) reviewing resumes, cover letters, and LinkedIn profiles from job seekers applying for software engineering, product, design, and marketing jobs.
Solved operates on a freemium model. By default, artifact reviews are free and publicly available, allowing the entire ecosystem to benefit from shared knowledge: open source. Users can upgrade for a small fee to expedite the review process / keep their information private.
Try it out and send any feedback our way.
Popularity: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
30. Show HN: Frosti – Access Management Copilot
URL: https://www.tryfrosti.com/beta
作者: kaistrubel
Could you please provide the text that needs to be translated?: Developers are overwhelmed by the hundreds of built in roles and thousands of permissions across AWS, GCP, and Azure. What if the developer was freed from the task of identifying the specific, least privileged role assignment or defaulting to the broad owner/contributor roles?
Frosti is an Access Management Copilot that will allow developers to request Just In Time access to the resources they need by describing their task in plain english rather than platform specific permissions. For example, tell Frosti “I need to rotate a secret in Azure Key Vault” and Frosti will respond “Request the Key Vault Secrets Officer role”. We imagine this becoming a Teams app which could generate a PIM request on the developers behalf, but for demo purposes we started with a simple standalone web app for role recommendations. Let us know what you think.
Try Frosti: https://www.tryfrosti.com/beta Demo Video: https://www.youtube.com/watch?v=XL5pOFjY7Pg
According to the MSFT’s State of Cloud Permission Risks Report, over 50% of identities are super admins, and only 1% of permissions granted to identities are actually used.
But who can blame the employees for over privilege? Here’s a scenario you see every day. You are working on a new project and need access to a resource, let’s say an Azure AI model. The resource owner has to pour over the IAM docs to pick the proper role out of the hundreds of built in roles. Often this first attempt fails (maybe they granted you access to the notebook but not the underlying cluster or data). Six iterations later you still don’t have the proper access, so they give up and add you as owner to the entire resource group and every relevant security group they can find. Since this works, no one bothers to later remove access and risk breaking your workflow.
So what’s the solution? To end the continuous chore of removing access, we need a simpler way to achieve “Zero Standing Privileges”. Rather than putting the onus on the developer to request the proper, time bound role, let them simply describe their task. This will simplify the process of picking a least privileged role and stop the common practice of developers defaulting to owner or contributor roles out of convenience.
受欢迎程度: 1 points | 2 comments
What do you think about this project? Share your thoughts in the comments!
31. Show HN: Misfits, a FITS table viewer for the terminal
URL: https://github.com/peppedilillo/misfits
作者: deppep
Please provide the text that you would like me to translate.: Hey any astrophysicist reading hn there? Was sick in august (yeah isn’t that nice) and made a small python fits table viewer for the terminal with textual. It would be great to have people try it so if you find any issue I can improve it. If you have no FITS file but would like to try regardless there are some test file in /test. Ciao!
受欢迎程度: 4 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
32. Show HN: The Multiverse Employee Handbook – Podcast
URL: https://multiverseemployeehandbook.com/
作者: taptak
Could you please provide the text that you would like me to translate?: Hey HN,
I recently launched a podcast called The Multiverse Employee Handbook, where I mix light-hearted humor with some science fiction and a sprinkle of quantum physics. It’s set in a surreal corporate universe where office politics, interdimensional HR dilemmas, and wormhole deliveries are all part of the daily grind.
Each narrative episode explores a different aspect of this multiverse, like quantum entanglement in the mailroom or cosmic deadlines. It’s definitely more on the fun and absurd side of things, but I try to weave in real science concepts (in the simplest way possible, of course).
Not trying to be the next big thing, just hoping a few folks out there might find it an entertaining break from reality. If you’re into sci-fi, comedy, or just need something light while you code, feel free to check it out. Would love to hear what you think—feedback is always welcome!
Here’s the link: multiverseemployeehandbook.com
Cheers!
人气: 5 points | 3 comments
What do you think about this project? Share your thoughts in the comments!
33. Show HN: Harp Proxy – open-source API Proxy for reliability and observability
作者: rdorgueil
人气: 7 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
34. Show HN: I proved myself wrong with a spreadsheet and made a Protein Calculator
URL: https://buzo.tools/protein-calculator/
作者: thatgurjot
受欢迎程度: 2 points | 1 comments
What do you think about this project? Share your thoughts in the comments!
35. Show HN: Discover most reviewed things.in/city
URL: https://things.in
作者: shyjal
It seems like you may have forgotten to include the text that needs to be translated. Could you please provide the content you’d like me to translate?: Hi HN,
When I used to travel to new cities, I relied on place ratings, but I noticed that sometimes, 5-star reviews were left by the owner and their family, which skewed the results.
I started trusting the number of reviews instead of the rating itself, but I found it difficult to find platforms that let me sort places this way.
I created this travel guide website to solve this issue. It features attractions, restaurants, and stays in various cities, all sorted by the number of reviews. Additionally, the website offers simple guidebooks for each city to help travelers plan their trips.
Here are examples for two cities:
San Francisco: https://things.in/sf Dubai: https://things.in/dubai
The site received a lot of love when I first launched it on Reddit: https://www.reddit.com/r/InternetIsBeautiful/comments/1fec8p…
I built this over a weekend without writing a single line of code using the Cursor editor. This is my third product built with Cursor. You can follow my product journey on X: x.com/shyjal.
Please give it a try and let me know what you think!
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
36. Show HN: Detective Stories – A 2-player “dark stories” game
URL: https://detective-stories.com/
作者: detectivestory
Popularity: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
37. Show HN: Finic – Open source platform for building browser automations
URL: https://github.com/finic-ai/finic
作者: jasonwcfan
Sure! Please provide the text that you would like me to translate.: Last year we launched a project called Psychic that did moderately well on hacker news, but was a commercial failure. We were able to find customers, but none with compelling and overlapping use cases. Everyone who was interested was too early to be a real customer.
This was our launch: https://news.ycombinator.com/item?id=36032081
We recently decided to revive and rebrand the project after seeing a sudden spike in interest from people who wanted to connect LLMs to data - but specifically through browsers. It’s also a problem we’ve experienced firsthand, having built scraping features into Psychic and previously working on bot detection at Robinhood.
If you haven’t built a web scraper or browser automation before, you might assume it’s very straightforward. People have been building scrapers for as long as the internet has existed, so there must be many tools for the job.
The truth is that web scraping strategies need to constantly adapt as web standard change, and as companies that don’t want to be scraped adopt new technologies to try and block it. The old standards never completely go away, so the longer the internet exists, the more edge cases you’ll need to account for. This adds up to a LOT of infrastructure that needs to be set up and a lot of schlep developers have to go through to get up and running.
Scraping is no easier today than it was 10 years ago - the problems are just different.
Finic is an open source platform for building and deploying browser agents. Browser agents are bots deployed to the cloud that mimic the behaviour of humans, like web scrapers or remote process automation (RPA) jobs. Simple examples include scripts that scrape static websites like the SEC’s EDGAR database. More complex use cases include integrating with legacy applications that don’t have public APIs, where the best way to automate data entry is to just manipulate HTML selectors (EHRs for example).
Our goal is to make Finic the easiest way to deploy a Playwright-based browser automation. With this launch, you can already do so in just 4 steps. Check out our docs for more info: https://docs.finic.io/quickstart
Popularity: 124 points | 67 comments
What do you think about this project? Share your thoughts in the comments!
38. Show HN: Opik, an open source LLM evaluation framework
URL: https://github.com/comet-ml/opik
作者: calebkaiser
Please provide the text you would like me to translate.: Hey HN! I’m Caleb, one of the contributors to Opik, a new open source framework for LLM evaluations.
Over the last few months, my colleagues and I have been working on a project to solve what we see as the most painful parts of writing evals for an LLM application. For this initial release, we’ve focused on a few core features that we think are the most essential:
- Simplifying the implementation of more complex LLM-based evaluation metrics, like Hallucination and Moderation.
- Enabling step-by-step tracking, such that you can test and debug each individual component of your LLM application, even in more complex multi-agent architectures.
- Exposing an API for “model unit tests” (built on Pytest), to allow you to run evals as part of your CI/CD pipelines
- Providing an easy UI for scoring, annotating, and versioning your logged LLM data, for further evaluation or training.
It’s often hard to feel like you can trust an LLM application in production, not just because of the stochastic nature of the model, but because of the opaqueness of the application itself. Our belief is that with better tooling for evaluations, we can meaningfully improve this situation, and unlock a new wave of LLM applications.
You can run Opik locally, or with a free API key via our cloud platform. You can use it with any model server or hosted model, but we currently have a built-in integration with the OpenAI Python library, which means it automatically works not just with OpenAI models, but with any model served via a compatible model server (ollama, vLLM, etc). Opik also currently has out-of-the-box integrations with LangChain, LlamaIndex, Ragas, and a few other popular tools.
This is our initial release of Opik, so if you have any feedback or questions, I’d love to hear them!
受欢迎程度: 74 points | 10 comments
What do you think about this project? Share your thoughts in the comments!
39. Show HN: Tinymind – Write and sync your blog and thoughts with GitHub
URL: https://github.com/mazzzystar/tinymind
作者: mazzystar
受欢迎程度: 7 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
40. Show HN: P5lab – a canvas-style UI for p5.js
URL: https://github.com/nicolaspasqualis/p5lab
作者: nnnnico
Sure! Please provide the text you would like to have translated.: I made this as a more serious playground for algorithmic design with p5.js after collecting some patterns/tools i’ve been using through my projects It’s basically a figma-style canvas but for running instances of p5.js code, with gui controllers and other utilities.
Been working on it for a couple of weeks, the code is messy, but it works ok for experimenting and I need to shift focus to other stuff for now.
Hope you find it interesting!
人气: 3 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
41. Show HN: Puck v0.16, the visual editor for React, introduces permissions (MIT)
URL: https://measured.co/blog/puck-v016
作者: chrisvxd
人气: 4 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
42. Show HN: Selectable – Postgres for Android
URL: https://getselectable.com/
作者: yen223
It seems like you’ve provided a single word, “Description.” Could you please provide more context or the text you would like me to translate?: I’ve built Selectable - a freemium Postgres client for Android devices. It’s currently available for open testing, links below.
The key feature is the SQL query editor. The query editor was designed to be mobile-friendly. With autocomplete and a touch-friendly way to select parts of a SQL query, writing SQL statements with the query editor can - in some cases - be faster than typing the same query on a desktop.
The app supports the usual SQL client things:
- Browsing schemas, tables, and columns
- Running custom SQL statements
- Viewing results as a list, or as a table.
- Exporting results to CSV
- Saving queries (You can save one query in the free version)
The project is in very early stages, and is still missing a lot of features - notably the ability to insert, update, delete, or perform DDL via the query editor.
The app currently only supports Postgres, and is currently only available on Android.
Please give it a spin, and let me know if you have any feedback or feature requests.
Website: https://getselectable.com/
Play store: https://play.google.com/store/apps/details?id=net.weiyen.sel…
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
43. Show HN: Sysm lets you play custom sounds on specified events
URL: https://github.com/jafarlihi/sysm
作者: jafarlihi
It seems that the text “Description” is too vague for translation. Could you please provide more context or details that you would like me to translate?: Reposting this because of recent Ask thread interest in a similar solution: https://news.ycombinator.com/item?id=41519046
Feel free to contribute sound pieces and new rules!
受欢迎程度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
44. Show HN: Run GPU Workloads Across Multiple Clouds
URL: https://github.com/beam-cloud/beta9
作者: Mernit
I’m sorry, but I can’t assist with that. If you have another request or need help with something specific, feel free to ask!: 4 points | 1 comments
What do you think about this project? Share your thoughts in the comments!
45. Show HN: Coding project to sift through Elon Musk’s Twitter feed for tech posts
URL: https://elons-technical-tweets.tech/
作者: michielt
Please provide the text you would like me to translate.: Hi! Fun little coding project I did to rate the tweets from Elon Musk on technical content, and retweet if they are above a certain threshold.
Rating is done using GPT-4o and Claude 3.5 Sonnet.
The number of political tweets from Musk was getting too much for me, but didn’t want to miss the technical stuff that he tweets about.
流行度: 1 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
46. Show HN: UI Replicator, clone any UI with only one click
URL: https://uireplicator.com/
作者: theandreibogdan
请提供您想要翻译的具体文本。: Clone any UI with only one click, and download the html + css of any website. Stop cloning UIs manually.
Popularity: 2 points | 9 comments
What do you think about this project? Share your thoughts in the comments!
47. Show HN: Local time of your friends in macOS menubar (native)
URL: https://there.pm
作者: morajabi
Please provide the text you would like me to translate.: Hey HN!
We got tired of looking at UTC offsets and remembering which of our friends or coworkers are in which time zone so we built There.
It lets you add “people” with names and photos instead of just time zones. It can fetch their photo by entering a Twitter (X) username too. Also, you can search by city names, countries, offsets or even abbreviations like PST.
It’s also open source! Written in Swift and Swift UI. Hope you enjoy it <3
Popularity: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
48. Show HN: A new kind of React store
URL: #
作者: ziolko
Please provide the text you would like translated, and I’d be happy to help!: Hey HN!
I am a developer on a break from employment. I want to use this time to recharge and work on projects that I always wanted to do but lacked energy after regular job.
One of the problems that we never solved properly is web application state management. What I want is a library that:
- encourages separating application state and logic from React components
- makes dealing async state dead simple (like react query)
- makes creating derived (computed) state dead simple (like computed properties in vue) including deriving async state
- allows to create reusable logic that can be imported as a library (like react hooks)
- allows to easily compose a few simple pieces of application state into a bigger logical unit that can be developed, consumed and tested separately
I’ve played with various ideas for the last few years, sometimes going crazy that I couldn’t find, or create something meeting all these criteria. Finally, a few months ago something clicked and that’s how active-store was born.
I am terrible at writing documentation, but I am good at writing code. Take a look at the store for a simple HN client: https://codesandbox.io/p/sandbox/headless-resonance-dfzgzw. It’s just ~100 lines of code with comments. The code handles the most important aspects of loading data from HN. I think this is a good example, because the HN API is quite difficult to work with.
There’s more documentation at https://github.com/ziolko/active-store and another project that I used as a playground at https://github.com/roombelt/timeline.
Despite the terrible documentation I encourage you to give it a try. I will be happy to answer your comments here.
受欢迎程度: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
49. Show HN: Codebased, an AI Search Engine for Code
URL: https://codebased.sh
作者: maxconradt
Please provide the text you’d like to have translated.: Codebased combines Tree Sitter for code awareness (find functions, data structures, constants, etc. not just lines of code), full-text search using SQLite, and semantic search using OpenAI embeddings + FAISS. Despite being implemented in Python, supporting semantic search, making multiple API calls for embedding and re-ranking, it is faster than ripgrep for runng searches against the Linux kernel (takes ~1 second vs. ~2 seconds, obviously depends on system, temperature, time of day, tidal forces, etc.) Up next:
- A Perplexity-like agent for interpreting results, making multiple follow-up searches, etc.
- Custom embedding and re-ranking stack
- Agent for running shell commands, editing code, etc. similar to SWe-Agent: https://arxiv.org/pdf/2405.15793.
人气: 16 points | 8 comments
What do you think about this project? Share your thoughts in the comments!
50. Show HN: DPaZ – Desktop Pan and Zoom
URL: https://github.com/sebmaynard/DPaZ
作者: sebmaynard
Could you please provide the text that you would like me to translate?: Hi HN!
I really like Miro and similar tools for whiteboarding - and one particularly nice feature of those is that you can pan and zoom around the whiteboard. Zoom into something to work on it, zoom out, pan a bit and find something else on a giant whiteboard.
This tool aims to replicate that experience for the Windows Desktop - it gives you the feeling of unlimited desktop real-estate, implemented with a relatively short Autohotkey script.
It’s a bit janky, but I use it daily; PRs welcome
人气: 6 points | 2 comments
What do you think about this project? Share your thoughts in the comments!
51. Show HN: Swift Gradient
URL: https://swiftgradient.toolkami.com
作者: SafeDusk
It seems like you didn’t provide any specific text to translate. Could you please share the text that you’d like to have translated?: Create beautiful gradients in seconds.
人气: 2 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
52. Show HN: Convert Any Webhook into Physical Print
URL: https://WebhookPrinter.com
作者: rubi1945
人气: 1 points | 1 comments
What do you think about this project? Share your thoughts in the comments!
53. Show HN: Open-Source FormSpree Alternative
作者: Oia1
人气: 4 points | 0 comments
What do you think about this project? Share your thoughts in the comments!
Conclusion
Today’s Show HN roundup showcases a diverse range of innovative projects. From AI-powered tools to creative coding solutions, these projects reflect the dynamic nature of our tech community. Which project caught your attention the most? Let us know in the comments!
Tags: #ShowHN #TechInnovation #DeveloperProjects #AI Applications #Open Source Software