Saturday, December 23, 2023

Show HN: Create and deploy multi-page web app prototypes using chat https://ift.tt/Jv7cQpn

Show HN: Create and deploy multi-page web app prototypes using chat gptengineer.app lets you prototype, deploy and iterate on web apps using plain english. Here is a video of how I use it: https://ift.tt/KvceVld We designed it around the principles: 1. Leverage LLMs where they work best: LLMs are faster than humans at putting together prototypes of a few hundred lines of code. By using them early on, and give high level change request commands when the project is small, and then let human devs take over, it’s a productivity boost. 2. non-technical + programmer collaboration: All changes from the app is made as git commits, so that non-technical people (via chat) and programmers are able to contribute to the same code. LLMs are expensive, but we left it open over the christmas holidays for anyone to try it out: https://ift.tt/fQ8H2AT Would love to hear what you think! https://ift.tt/vSDoW9c December 22, 2023 at 11:28PM

Friday, December 22, 2023

Show HN: Local fine tuning for Mistral and SDXL, GPU mem/latency optimization https://ift.tt/cmi1pFw

Show HN: Local fine tuning for Mistral and SDXL, GPU mem/latency optimization 100% bootstrapped new startup. It lets you fine tune Mistral-7B and SDXL. In particular, for the LLM fine tuning we implemented a dataprep pipeline that turns websites/pdfs/doc files into question-answer pairs for training the small LLM using an big LLM. It includes a GPU scheduler that can do finegrained GPU memory scheduling (Kubernetes can only do whole-GPU, we do it per-GB of GPU memory to pack both inference and fine tuning jobs into the same fleet) to fit model instances into GPU memory to optimally trade off user facing latency with GPU memory utilization It's a pretty simple stack of control plane and a fat container that runs anywhere you can get hold of a GPU (e.g. runpod). Architecture: https://ift.tt/PLn24JV Demo walkthrough showing runner dashboard: https://ift.tt/WC6Pl8h Run it yourself: https://ift.tt/IE29qWv Discord: https://ift.tt/wmZpdXB Please roast me! https://ift.tt/WC6Pl8h December 22, 2023 at 01:43AM

How Your Feedback is Shaping the Bayview Community Shuttle

How Your Feedback is Shaping the Bayview Community Shuttle
By Hana O’Neill

A new community shuttle is coming to Bayview-Hunters Point in 2024, and we need your help designing it! The SFMTA is partnering with the California Air Resources Board and nine community-based organizations to launch a dynamic service community shuttle that truly meets your needs. The shuttle is funded through California Climate Investments, a statewide initiative that puts billions of Cap-and-Trade dollars to work reducing greenhouse gas emissions, strengthening the economy and improving public health and the environment — particularly in disadvantaged communities. 

A “dynamic service” shuttle is one that doesn’t have a fixed route like a bus line, but instead operates in a “service area.” The shuttle will pick up and drop off riders throughout the service area, which can help you reach destinations that were previously hard to get to. The Bayview is a geographically isolated community and the community identified the need for a shuttle like this in the Bayview Community Based Transportation Plan

We’re currently working on defining that service area. We’ve held 28 events where we talked to almost 2,000 residents, but we know not everyone can come to community meetings or other events. So we also conducted a survey that residents filled out online or on paper at outreach events. We surveyed 629 residents, who gave us 1,060 places they wanted to take a shuttle to. Here’s a summary of the results: 

Many people wanted to take short trips that start and end within the Bayview. 76 people told us that they wanted connections to specific transit lines, such as to the T Third. Others gave destinations that are easily accessible via already-existing Muni lines, such as Chinatown. While a shuttle may not be able to take you all the way to that destination, a direct lift to the T Third would make that trip much easier. 

A legend for a map of San Francisco displaying Muni Routes that serve the Bayview and frequency of survey responses. Most survey responses are in the Bayview or along Muni routes. Frequently requested destinations include Chinatown, Downtown, 24th St BART, the San Bruno Ave commercial corridor, and SF General Hospital.A map of San Francisco displaying Muni Routes that serve the bayview and frequency of survey responses. Most survey responses are in the Bayview or along Muni routes. Frequently requested destinations include Chinatown, Downtown, 24th St BART, the San Bruno Ave commercial corridor, and SF General Hospital.

Desired destination of survey respondents.

 

Initial Service Plan/Alternative A: 

The initial service plan that we presented to residents included service from 7 a.m. to 7 p.m., seven days a week. Riders would wait up to 15 minutes for a ride to Bayview destinations, as well as a few others outside of the neighborhood, such as the 22nd Street Caltrain station, 24th Street BART station and Zuckerberg San Francisco General Hospital.  

The initial concept shows a map of the service area including the Bayview itself, and 3 potential external destinations: 24th St BART, 22nd St Caltrain, and SF General Hospital. This alternative would have approximately 12 hours of service 7 days a week, a moderately large service area, and moderate wait timesA chart depicting the service hours related to the Alternative A map: this alternative would have approximately 12 hours of service 7 days a week, a moderately large service area, and moderate wait times

Alternative A: Initial conception of service plan.

 

Alternative B: 

How does this plan sound to you? Would you rather have a larger service area for the shuttle so that you could be picked up or dropped off in neighboring areas (shown in gray in the Alternative B map)? This alternative would require running service for a shorter time period – for instance, operating only on weekdays – and with longer wait times. Is this tradeoff worth it? We’d love to hear from you. 

The larger service area alternative B shows a map of the service area including the Bayview itself, and many neighboring areas, potentially including Potrero Hill, the Mission, and parts of the Portola and Visitacion Valley, though exact borders have not been determined. This alternative would have approximately 12 hours of service 5 days a week, a large service area and longer wait times.A graphic chart depicting the service hours related to the Alternative B map. This alternative would have approximately 12 hours of service 5 days a week, a large service area, and longer wait times.

Alternative B: Larger service area and tradeoffs

 

Alternative C: 

We could also run the shuttle for a longer time period, such as from 6 a.m. to 10 p.m., seven days a week, with short wait times – but this would require focusing on local neighborhood connections. The shuttle would serve only the Bayview, with no destinations outside of the neighborhood. 

Alternative C shows a map of the service area including immediate neighborhood connections within the Bayview. This alternative would allow us to run approximately 16 hours of service 7 days per week, and allow shorter wait times, but leaves a smaller service area. A chart reflecting the more service hours found in the Alternative C map of the service area including immediate neighborhood connections within the Bayview. This alternative would allow us to run approximately 16 hours of service 7 days per week, and allow shorter wait times, but leaves a smaller service area.

Alternative C: More service hours and tradeoffs

 

Would you rather be able to go more places, even if that means you wait longer, and the shuttle doesn’t run for as many hours of the day? Or would you rather get a ride more quickly, at more times of day, even if it can’t take you directly to as many locations? Please let us know what you think! 

You can sign up for project updates here to learn about upcoming meetings, including our Community Congress on January 24th. You also can email us at BayviewShuttle@sfmta.com.  

As we continue to do outreach to hear about your priorities, we’re also working to select a company to run this shuttle service. When all is said and done, we’re hoping to start service by the middle of 2024. 



Published December 22, 2023 at 12:50AM
https://ift.tt/0LUISo4

Show HN: I'm open-sourcing my game engine https://ift.tt/VBPJtYQ

Show HN: I'm open-sourcing my game engine Modd.io is a collaborative game editor that runs in browser. It's kind of like Figma for game dev. We made this engine low-code and multiplayer-first, so developeres can quickly prototype casual multiplayer games. I hope some of you guys will find this useful. Would love to hear feedback also. Thank you. Engine Demo: https://www.modd.io https://ift.tt/m1QwthR December 20, 2023 at 01:08PM

Thursday, December 21, 2023

Show HN: fx-upscale – Metal-powered spatial video upscaling https://ift.tt/h8T3NJe

Show HN: fx-upscale – Metal-powered spatial video upscaling Hi! This is a simple CLI I've created for upscaling videos using Apple's MetalFX ( https://ift.tt/FEbLDG7 ). There are a few improvements I plan to make, such as HDR support and a progress bar, but if it sounds interesting, please give it a try and let me know what you think! https://ift.tt/Zb7aIDV December 21, 2023 at 05:28AM

Show HN: JungleReader – A free and open-source web reader https://ift.tt/VYar9Bg

Show HN: JungleReader – A free and open-source web reader Hello! I built a web-app called JungleReader that lets you follow RSS, Mastodon, YouTube, Reddit, and random blogs, all in one place. I made JungleReader as a central place to follow the blogs and sites I like, which I otherwise often lose track of. My next project (which is WIP) is a web-app called JungleWriter where you can build a simple blog and export it for hosting. I love finding a good, free-standing blog, and hope to make it easier for people to read and write them. Feedback is appreciated, thanks :) https://ift.tt/6VterGs December 21, 2023 at 03:45AM

Show HN: Slack knowledge curator to extract FAQ from discussion threads https://ift.tt/P1wbXlI

Show HN: Slack knowledge curator to extract FAQ from discussion threads Hi HN! Excited to share Snowshoe, a Slack bot that transforms your Slack threads into a searchable StackOverflow-style Q&A. Snowshoe automatically curates key discussions from Slack, creating a dynamic, easy-to-navigate knowledge hub. A lot of technical discussions among engineers happen in Slack threads. Yet the unstructured thread makes its inherent knowledge difficult for search and thus the discussion insights aren’t often well documented & maintained. LLM is surprisingly good at summarization & extraction. Thus I think it’s good timing to give it a try. Currently Snowshoe focuses on knowledge extraction in Q&A format. What’s next? I feel like there would be a lot of value in the auto-curated Q&A, e.g. auto question answering, knowledge hub, Slack archive for SEO. Happy to hear your thoughts! https://snowshoe.dev/ December 21, 2023 at 12:21AM

Show HN: Do You Know RGB? https://ift.tt/t8kUpbO

Show HN: Do You Know RGB? https://ift.tt/OWhvmMT June 24, 2025 at 01:49PM