Show HN: Beating OpenAI's structured outputs on cost, accuracy and speed https://ift.tt/J1ZBcXt

Show HN: Beating OpenAI's structured outputs on cost, accuracy and speed https://ift.tt/yiXoS9I August 15, 2024 at 12:32AM

Show HN: SlateDB – An embedded storage engine built on object storage https://ift.tt/HszOt9w

Show HN: SlateDB – An embedded storage engine built on object storage SlateDB is an embedded storage engine built as a log-structured merge-tree. Unlike traditional LSM-tree storage engines, SlateDB writes data to object storage (S3, GCS, ABS, MinIO, Tigris, and so on). Leveraging object storage allows SlateDB to provide bottomless storage capacity, high durability, and easy replication. The trade-off is that object storage has a higher latency and higher API cost than local disk. To mitigate high write API costs (PUTs), SlateDB batches writes. Rather than writing every put() call to object storage, MemTables are flushed periodically to object storage as a string-sorted table (SST). The flush interval is configurable. To mitigate write latency, SlateDB provides an async put method. Clients that prefer strong durability can await on put until the MemTable is flushed to object storage (trading latency for durability). Clients that prefer lower latency can simply ignore the future returned by put. To mitigate read latency and read API costs (GETs), SlateDB will use standard LSM-tree caching techniques: in-memory block caches, compression, bloom filters, and local SST disk caches. https://ift.tt/rUX9C2l August 15, 2024 at 12:47AM

Show HN: AnimeGenAi – AI-powered anime style image and video generator https://ift.tt/bOahlC3

Show HN: AnimeGenAi – AI-powered anime style image and video generator https://animegenai.com August 14, 2024 at 01:42AM

Show HN: AI Bartender in a Virtual Bar https://ift.tt/aNfqLSj

Show HN: AI Bartender in a Virtual Bar Note: The avatar will talk to you, so make sure you check your volume! Hello HN! I've seen a bunch of AI character type stuff online, but all of them were boring chat interfaces. I thought it would be more fun to give the characters an avatar and a 3d environment to interact with. The stack I'm using is Claude 3.5 for the LLM, OpenAI TTS, Stable Diffusion for generating drinks, and three.js for rendering. I exposed the prompt I'm using so people can play around with it by clicking the robot icon. If people find it interesting, I might add more environments, character customization options, etc. Hope you enjoy and let me know your thoughts/questions in the comments! If you want to know more, follow me on Twitter or join the Discord! https://ift.tt/opsq038 https://x.com/neelmango https://ift.tt/DjiZ42F August 13, 2024 at 10:58PM

Modernizing Potrero Yard: Give Feedback on Public Art

Modernizing Potrero Yard: Give Feedback on Public Art
By Caroline Cabral

Photo of Potrero Yard as it looks today. The Potrero Yard Modernization Project plans to replace the 100-year-old bus yard with a modern bus facility. The new facility will include large-scale public art, and you have a chance to share feedback on the artist-proposed designs. The San Francisco Arts Commission manages public art for this project. They invite you to comment on conceptual designs by the finalist artists or artist teams for two glass art installations for the site. Share your Feedback You can view the art proposals for each location on the facility and share comments on two...

Published August 13, 2024 at 05:30AM
https://ift.tt/RzYwbSo

Show HN: AI co-worker for system software development (Rust,C,C++,pdf) https://ift.tt/fW0adYz

Show HN: AI co-worker for system software development (Rust,C,C++,pdf) Hey Everybody, We are really excited to release the 1st version of H2LooP studio today. https://h2loop.ai/ H2LooP Studio helps system software engineers generate code from technical specs, debug issues, and understand complex code in C, C++, Go, and Rust. Under the hood, it uses the H2LooP Data Engine to create instruction-tuned datasets from data sheets and source code. Models are what they eat. We create high-quality, pre-vetted domain-specific training data (telecom, IoT, automotive, consumer electronics) at scale for fine-tuning small language models. We leverage both LLMs and human expertise (system knowledge) to build this dataset. Why are we building H2Loop? 1.Challenges in System Code: -System code presents significant challenges for LLMs that lack specialised pre-training. -Existing tools like GitHub Copilot struggle with tasks such as generating device driver code, debugging network kernel crashes, and interpreting hardware schematics. 2.Limitations of Current Coding Assistants: -Results from generic coding assistants are often unclear and insufficient. -These tools are unable to handle technical specifications or crash logs, which are essential for system software development. -System developers frequently need to reference specifications like Wi-Fi, Bluetooth, or network protocols while coding, but current tools fail to meet these needs. 3.Specialised Requirements for System Software: -System software is typically written in languages like C, C++, Go, and Rust, often in closed-source projects. -Enterprises need specialised solutions that understand their specific domain and coding standards. Challenges in Generating Accurate Code from Technical Specifications: 1.Unstructured Format of Technical Specifications: -Technical specifications are often in PDF format, which is inherently unstructured. -Parsing PDFs that include images, tables, and various text elements, and aligning them with reference sample code, presents a significant challenge. 2.Difficulty in Creating Domain-Specific Datasets: -Developing a question-and-answer coding dataset for specialised domains like automotive or telecom, suitable for LLM training, is a complex task. 3.Necessity of Expert Review: -Expert review of the training dataset is crucial. For example, if a dataset is created for socket creation in a networking protocol, it must be meticulously checked by an expert before being used for fine-tuning. The Solution: 1.RAG-Based Parsing and Chunking: -We employ a Retrieval-Augmented Generation (RAG) solution to parse and chunk PDFs effectively. -By combining LLM and manual methods, we align the content from PDFs with source code to create an instruction tuned dataset. 2.Expert Review and Validation: -Our team of system and domain experts thoroughly review and validate the training datasets, which are formatted in JSON. 3.Collaborative Fine-Tuning: -We partner with enterprises to transform their code and technical specifications into expert-vetted, domain-specific datasets. -We then assist in fine-tuning a small language model tailored to their domain and coding standards. Who can use H2LooP: H2LooP is a valuable tool for professionals like developers, product managers, and CTOs. If you're working on proprietary software, frequently coding from technical specifications,H2LooP is for you. Demo: https://ift.tt/TbtJBM3 H2LooP Studio is hosted in the cloud. You can download sample technical specifications and experiment with the H2LooP model to generate system software code. We will soon be releasing the H2LooP Data Engine, which will allow you to create training datasets by uploading code and PDFs. For more details, refer to https://ift.tt/LBo5NCb Also please join our community at : - Slack : https://ift.tt/hUaOo1m - Twitter : https://x.com/h2loopinc Would love to hear your feedback & how we can make this better. Thank you, Team H2LooP https://h2loop.ai/ August 13, 2024 at 09:02PM

Show HN: Llmq – Pipe stuff in the shell to AI https://ift.tt/tbvjiMY

Show HN: Llmq – Pipe stuff in the shell to AI I just put this together real quick and am finding it useful. Sharing in case anyone else would benefit from it. https://ift.tt/Q9i7rxK August 12, 2024 at 11:12PM