Show HN: HTML for People https://ift.tt/Mng0Zvp

Show HN: HTML for People https://ift.tt/cV1IBYv October 10, 2024 at 11:17PM

Show HN: Arch – an intelligent prompt gateway built on Envoy https://ift.tt/we0i1jy

Show HN: Arch – an intelligent prompt gateway built on Envoy Hi HN! My name is Adil Hafeez, and I am the Co-Founder at Katanemo and the lead developer behind Arch - an open source project for developers to build faster, generative AI apps. Previously I worked on Envoy at Lyft. Engineered with purpose-built LLMs, Arch handles the critical but undifferentiated tasks related to the handling and processing of prompts, including detecting and rejecting jailbreak attempts, intelligently calling “backend” APIs to fulfill the user’s request represented in a prompt, routing to and offering disaster recovery between upstream LLMs, and managing the observability of prompts and LLM interactions in a centralized way - all outside business logic. Here are some additional key details of the project, * Built on top of Envoy and is written in rust. It runs alongside application servers, and uses Envoy's proven HTTP management and scalability features to handle traffic related to prompts and LLMs. * Function calling for fast agentic and RAG apps. Engineered with purpose-built fast LLMs to handle fast, cost-effective, and accurate prompt-based tasks like function/API calling, and parameter extraction from prompts. * Prompt guardrails to prevent jailbreak attempts and ensure safe user interactions without writing a single line of code. * Manages LLM calls, offering smart retries, automatic cutover, and resilient upstream connections for continuous availability. * Uses the W3C Trace Context standard to enable complete request tracing across applications, ensuring compatibility with observability tools, and provides metrics to monitor latency, token usage, and error rates, helping optimize AI application performance. This is our first release, and would love to build alongside the community. We are just getting started on reinventing what we could do at the networking layer for prompts. Do check it out on GitHub at https://ift.tt/6qDuh81 . Please leave a comment or feedback here and I will be happy to answer. (I did make a mistake earlier not properly tagging the title with Show HN and the post didn't end up in Show HN. Later I updated the title but then my text started showing up as a comment. So I redid the post again, sorry if you had looked at the previous post already. Here is the link to the old post https://ift.tt/sh3n1JT ) https://ift.tt/DK6lNBt October 10, 2024 at 11:15PM

Show HN: AI detects resume "red flags" https://ift.tt/nRjzCWk

Show HN: AI detects resume "red flags" https://ift.tt/Pf1bVHr October 10, 2024 at 01:53AM

Show HN: VRE Dataset generation for MultiTask vision models training from videos https://ift.tt/5UJXDq7

Show HN: VRE Dataset generation for MultiTask vision models training from videos Been working on this tool for my PhD which involves training multi task vision models using various pre-trained models as inputs or pseudolabels in order to improve generalization. I work mostly on UAV datasets, but it should work okay on indoor scenes or self driving (at least Marigold and Mask2Former). For example, this dataset was generated using this tool: https://ift.tt/MAit1kz I'm quite aggressively trying to "just get the nn.Module" from the public repos that other researchers put up in their overly convoluted frameworks. A simple `forward(rgb_input: torch.Tensor) -> torch.Tensor` is nice, having 100 imports from a generic framework that has versions incompatibilities with everything else is not. PS: most mains are standalone runnable too, i.e. - https://ift.tt/vVDinUY... or - https://ift.tt/vVDinUY... https://ift.tt/yi194cs October 9, 2024 at 11:09PM

Show HN: GodmodeHQ – Using AI agents to find public info on any company https://ift.tt/9eUpt3l

Show HN: GodmodeHQ – Using AI agents to find public info on any company Hi hackers, Mert from GodmodeHQ here ( https://godmodehq.com/ ) We are building a prospecting platform for deep research so anyone can understand their prospects deeply and send better outreach instead of spray and praying a generic message. Sales landscape is pretty crowded. Most solutions focus on sending the same email or Linkedin message to thousands of people. This approach worked between 2010 to 2020 at best, however it became such a big habit that everyone started doing it. When you receive an outreach email that talks about the solution they offer without inferring how the solution could solve your problem, you automatically mark it as spam. Now with AI, people started personalising. However this personalisation goes to the extent of where someone graduated from or if they shared a pic of eating ice cream on Instagram. This makes the outreach personal but still useless. At GodmodeHQ, we implemented AI agents in form of columns in a spreadsheet. This resembles Perplexity integrated into a spreadsheet. The agents recognise your website and embeds your ICP and context. So the only thing you need to do is ask: "What are some of the pain points we can solve for the company?" "Does the company have SOC Type 2 certification?" "What are the recent feature launches they did?" and more. The agent automatically starts filling the information for you. Then you can use this information to really understand what problems you can solve for the customer. Last step is reaching out to the customer using email and Linkedin. You can automate this part in GodmodeHQ as well however we do not deal with domain warmups or other similar practices that aim to let you send 10k emails on a day. By design we encourage the users to go through the information agents have uncovered. Our belief is that everyone sells. Whether it is your code or simple ecommerce products, you are a seller. You need to understand how you can sell. Our vision is to turn non-salespeople into successful salespeople using the help of AI. We hope you enjoy GodmodeHQ as much as we do. I would love to hear your feedback. Best, Mert https://godmodehq.com/ October 9, 2024 at 11:36PM

Show HN: The Repairable Infinite Battery is now live on Indiegogo https://ift.tt/JL9EGqH

Show HN: The Repairable Infinite Battery is now live on Indiegogo Thanks to this community for continued support! We've been able to bring our prototype Gouach Infinite Battery — Repairable and Fireproof — to production and we're live on Indiegogo! For those who are interested in buying a battery, there's a 25% off early-bird offer! For those who just want to chat, feel free to ask any question and I'll answer! https://ift.tt/8UhR2Fr October 9, 2024 at 12:23AM

Show HN: Winamp and other media players, rebuilt for the web with Web Components https://ift.tt/xVGSHKm

Show HN: Winamp and other media players, rebuilt for the web with Web Components https://player.style October 8, 2024 at 11:57PM