Tenjin (YC S14) Is Hiring a Senior Ad Attribution Engineer (Ruby, Go)
1 by amirmanji | 0 comments on Hacker News.
1 by amirmanji | 0 comments on Hacker News.
Show HN: Colanode, open-source and local-first Slack and Notion alternative
10 by hakanshehu | 0 comments on Hacker News.
Hey HN, I'm Hakan, the founder of Colanode ( https://ift.tt/u0kBQie ), an open-source, local-first collaboration app combining the best of Slack-style chats and Notion-style note-taking, fully self-hostable for complete data control. Here's a quick demo: https://www.youtube.com/watch?v=wp1hoSCEArg As a heavy Notion user, I often found it tough to get my teams fully onboard since people naturally gravitate toward chat for quick interactions. Maintaining context between chat apps like Slack and documentation apps like Notion became increasingly frustrating. Switching contexts, losing track of information, and managing data across multiple tools created unnecessary friction. This frustration led me to build Colanode, a single platform integrating structured notes and knowledge management with real-time chat. After building the first version, early feedback highlighted a critical issue: teams/organizations want full control over sensitive data, especially conversations. That's why I decided to open-source Colanode under an Apache 2.0 license, making it fully self-hostable so you can retain complete ownership and privacy over your data. Colanode is built with simplicity and extensibility in mind, using only open-source tools and avoiding any vendor or cloud lock-in. It features a local-first architecture offering complete offline support. From a technical perspective, Colanode consists of a Node.js server API and an Electron desktop client, with mobile apps coming soon. Everything in Colanode is represented as a node (e.g., message, file, folder, chat, channel, database, record), each with specific attributes and permissions. All reads and writes performed by the desktop client happen locally within a SQLite database, and changes sync seamlessly via a synchronization engine built on top of SQLite, Postgres, and Yjs—a CRDT library for conflict resolution. The server then propagates these changes to other collaborators. You can self-host the server in any environment using Docker, Postgres, Redis, and any S3-compatible storage, and connect using the official desktop client, which supports simultaneous connections to multiple servers and accounts. This local-first approach also prepares us for future integrations with fully local LLMs, further enhancing privacy and performance. I'd love your feedback and suggestions on Colanode. What features would you like to see? What would you change? Thanks, looking forward to your thoughts!
10 by hakanshehu | 0 comments on Hacker News.
Hey HN, I'm Hakan, the founder of Colanode ( https://ift.tt/u0kBQie ), an open-source, local-first collaboration app combining the best of Slack-style chats and Notion-style note-taking, fully self-hostable for complete data control. Here's a quick demo: https://www.youtube.com/watch?v=wp1hoSCEArg As a heavy Notion user, I often found it tough to get my teams fully onboard since people naturally gravitate toward chat for quick interactions. Maintaining context between chat apps like Slack and documentation apps like Notion became increasingly frustrating. Switching contexts, losing track of information, and managing data across multiple tools created unnecessary friction. This frustration led me to build Colanode, a single platform integrating structured notes and knowledge management with real-time chat. After building the first version, early feedback highlighted a critical issue: teams/organizations want full control over sensitive data, especially conversations. That's why I decided to open-source Colanode under an Apache 2.0 license, making it fully self-hostable so you can retain complete ownership and privacy over your data. Colanode is built with simplicity and extensibility in mind, using only open-source tools and avoiding any vendor or cloud lock-in. It features a local-first architecture offering complete offline support. From a technical perspective, Colanode consists of a Node.js server API and an Electron desktop client, with mobile apps coming soon. Everything in Colanode is represented as a node (e.g., message, file, folder, chat, channel, database, record), each with specific attributes and permissions. All reads and writes performed by the desktop client happen locally within a SQLite database, and changes sync seamlessly via a synchronization engine built on top of SQLite, Postgres, and Yjs—a CRDT library for conflict resolution. The server then propagates these changes to other collaborators. You can self-host the server in any environment using Docker, Postgres, Redis, and any S3-compatible storage, and connect using the official desktop client, which supports simultaneous connections to multiple servers and accounts. This local-first approach also prepares us for future integrations with fully local LLMs, further enhancing privacy and performance. I'd love your feedback and suggestions on Colanode. What features would you like to see? What would you change? Thanks, looking forward to your thoughts!
EU Energy labelling will apply to phones and tablets from June 2025
52 by sohkamyung | 23 comments on Hacker News.
52 by sohkamyung | 23 comments on Hacker News.
Ask HN: Share your AI prompt that stumps every model
4 by owendarko | 1 comments on Hacker News.
I had an idea for creating a crowdsourced database of AI prompts that no AI model could yet crack. I've seen a bunch of those prompts scattered across HN, so thought to open a thread here so we can maybe have a centralied location for this. Share your prompt that stumps every AI model here.
4 by owendarko | 1 comments on Hacker News.
I had an idea for creating a crowdsourced database of AI prompts that no AI model could yet crack. I've seen a bunch of those prompts scattered across HN, so thought to open a thread here so we can maybe have a centralied location for this. Share your prompt that stumps every AI model here.
A Principled Approach to Querying Data – A Type-Safe Search DSL
15 by kioku | 2 comments on Hacker News.
15 by kioku | 2 comments on Hacker News.
Asymmetric Content Moderation in Search Markets: The Case of Adult Websites
8 by amadeuspagel | 0 comments on Hacker News.
8 by amadeuspagel | 0 comments on Hacker News.
Show HN: Lemon Slice Live, a real-time video-audio AI model
7 by lcolucci | 3 comments on Hacker News.
Hey HN, this is Lina, Andrew, and Sidney from Lemon Slice. We’ve trained a custom diffusion transformer (DiT) model that achieves video streaming at 25fps and wrapped it into a demo that allows anyone to turn a photo into a real-time, talking avatar. Here’s an example conversation from co-founder Andrew: https://www.youtube.com/watch?v=CeYp5xQMFZY . Try it for yourself at: https://ift.tt/cvgCU4b . (Btw, we used to be called Infinity AI and did a Show HN under that name last year: https://ift.tt/xn972ag .) Unlike existing avatar video chat platforms like HeyGen, Tolan, or Apple Memoji filters, we do not require training custom models, rigging a character ahead of time, or having a human drive the avatar. Our tech allows users to create and immediately video-call a custom character by uploading a single image. The character image can be any style - from photorealistic to cartoons, paintings, and more. To achieve this demo, we had to do the following (among other things! but these were the hardest): 1. Training a fast DiT model. To make our video generation fast, we had to both design a model that made the right trade-offs between speed and quality, and use standard distillation approaches. We first trained a custom video diffusion transformer (DiT) from scratch that achieves excellent lip and facial expression sync to audio. To further optimize the model for speed, we applied teacher-student distillation. The distilled model achieves 25fps video generation at 256-px resolution. Purpose-built transformer ASICs will eventually allow us to stream our video model at 4k resolution. 2. Solving the infinite video problem. Most video DiT models (Sora, Runway, Kling) generate 5-second chunks. They can iteratively extend it by another 5sec by feeding the end of the 1st chunk into the start of the 2nd in an autoregressive manner. Unfortunately the models experience quality degradation after multiple extensions due to accumulation of generation errors. We developed a temporal consistency preservation technique that maintains visual coherence across long sequences. Our technique significantly reduces artifact accumulation and allows us to generate indefinitely-long videos. 3. A complex streaming architecture with minimal latency. Enabling an end-to-end avatar zoom call requires several building blocks, including voice transcription, LLM inference, and text-to-speech generation in addition to video generation. We use Deepgram as our AI voice partner. Modal as the end-to-end compute platform. And Daily.co and Pipecat to help build a parallel processing pipeline that orchestrates everything via continuously streaming chunks. Our system achieves end-to-end latency of 3-6 seconds from user input to avatar response. Our target is <2 second latency. More technical details here: https://lemonslice.com/live/technical-report . Current limitations that we want to solve include: (1) enabling whole-body and background motions (we’re training a next-gen model for this), (2) reducing delays and improving resolution (purpose-built ASICs will help), (3) training a model on dyadic conversations so that avatars learn to listen naturally, and (4) allowing the character to “see you” and respond to what they see to create a more natural and engaging conversation. We believe that generative video will usher in a new media type centered around interactivity: TV shows, movies, ads, and online courses will stop and talk to us. Our entertainment will be a mixture of passive and active experiences depending on what we’re in the mood for. Well, prediction is hard, especially about the future, but that’s how we see it anyway! We’d love for you to try out the demo and let us know what you think! Post your characters and/or conversation recordings below.
7 by lcolucci | 3 comments on Hacker News.
Hey HN, this is Lina, Andrew, and Sidney from Lemon Slice. We’ve trained a custom diffusion transformer (DiT) model that achieves video streaming at 25fps and wrapped it into a demo that allows anyone to turn a photo into a real-time, talking avatar. Here’s an example conversation from co-founder Andrew: https://www.youtube.com/watch?v=CeYp5xQMFZY . Try it for yourself at: https://ift.tt/cvgCU4b . (Btw, we used to be called Infinity AI and did a Show HN under that name last year: https://ift.tt/xn972ag .) Unlike existing avatar video chat platforms like HeyGen, Tolan, or Apple Memoji filters, we do not require training custom models, rigging a character ahead of time, or having a human drive the avatar. Our tech allows users to create and immediately video-call a custom character by uploading a single image. The character image can be any style - from photorealistic to cartoons, paintings, and more. To achieve this demo, we had to do the following (among other things! but these were the hardest): 1. Training a fast DiT model. To make our video generation fast, we had to both design a model that made the right trade-offs between speed and quality, and use standard distillation approaches. We first trained a custom video diffusion transformer (DiT) from scratch that achieves excellent lip and facial expression sync to audio. To further optimize the model for speed, we applied teacher-student distillation. The distilled model achieves 25fps video generation at 256-px resolution. Purpose-built transformer ASICs will eventually allow us to stream our video model at 4k resolution. 2. Solving the infinite video problem. Most video DiT models (Sora, Runway, Kling) generate 5-second chunks. They can iteratively extend it by another 5sec by feeding the end of the 1st chunk into the start of the 2nd in an autoregressive manner. Unfortunately the models experience quality degradation after multiple extensions due to accumulation of generation errors. We developed a temporal consistency preservation technique that maintains visual coherence across long sequences. Our technique significantly reduces artifact accumulation and allows us to generate indefinitely-long videos. 3. A complex streaming architecture with minimal latency. Enabling an end-to-end avatar zoom call requires several building blocks, including voice transcription, LLM inference, and text-to-speech generation in addition to video generation. We use Deepgram as our AI voice partner. Modal as the end-to-end compute platform. And Daily.co and Pipecat to help build a parallel processing pipeline that orchestrates everything via continuously streaming chunks. Our system achieves end-to-end latency of 3-6 seconds from user input to avatar response. Our target is <2 second latency. More technical details here: https://lemonslice.com/live/technical-report . Current limitations that we want to solve include: (1) enabling whole-body and background motions (we’re training a next-gen model for this), (2) reducing delays and improving resolution (purpose-built ASICs will help), (3) training a model on dyadic conversations so that avatars learn to listen naturally, and (4) allowing the character to “see you” and respond to what they see to create a more natural and engaging conversation. We believe that generative video will usher in a new media type centered around interactivity: TV shows, movies, ads, and online courses will stop and talk to us. Our entertainment will be a mixture of passive and active experiences depending on what we’re in the mood for. Well, prediction is hard, especially about the future, but that’s how we see it anyway! We’d love for you to try out the demo and let us know what you think! Post your characters and/or conversation recordings below.