Mobavenue Launches Neural Engine As An AI Intelligence Layer For MarketersMobavenue has launched the Mobavenue Neural Engine, an AI intelligence layer progressively being embedded across its advertising platforms ...
Enterprise AI agents are only as reliable as the messiest documents behind themEnterprise AI has largely been built around context engineering. Teams connect enterprise systems, generate chunks and embeddings, build retrieval pipelines, and assemble the context needed by ...
GPU Board For Rugged Edge SystemsA compact VPX computing board combines GPU acceleration, high-speed memory and rugged thermal options for demanding embedded workloads at the edge.
Why Trusted Clinical AI Is Healthcare's Next Cybersecurity ImperativeAs AI becomes embedded in clinical workflows, a new risk is emerging that existing cybersecurity frameworks weren't designed to address.
Show HN: Youtilitics, energy usage analytics from utility data, no hardwareI posted an early version of this 2 years ago. Rebuilt most of it since then, so posting again with what's actually different.Youtilitics still pulls interval data your utility already collects (Green Button/ESPI — 15-60 min whole-house kWh) and turns it into usage analytics. The analytics stack is different: baseline anomaly detection (median/MAD by time-of-day bucket, not a flat threshold), overnight step-change alerts, usage heatmaps, cycle-over-cycle comparisons, etc. It is still based on a model trained on human-labelled data: used to differentiate baseline (eg vampire load) vs EV vs HVAC vs the rest. Inference runs with a python script when new data is received from the utility.I also changed the mobile app. It is still mainly used to store the user credentials without us having them (they never leave the device), for security/privacy reasons. This is used to scrape usage data from the utility dashboard, when they don't support Green Button. However now it also runs an embedded P
Show HN: Building Table Canon, an AI Campaign Memory Engine for TTRPGsHey HN! I built Table Canon to solve a problem my playgroup kept running into: 3-4 hour tabletop gaming sessions leave behind massive audio recordings, but standard meeting note-takers treat every session as an isolated island, butcher fantasy terms, and don't know who is speaking.I wanted an engine that tracks long-term state across months of games, so I built a pipeline to extract entity updates, open quest hooks, and character promises across sessions.The Tech Stack:* Transcription: whisper-large-v3-turbo * Diarization: pyannote for speaker embeddings & voice profile matching Extraction & Memory: OpenAI API with Structured Outputs (JSON Schema enforcement for state updates) * TTS & Audio Recaps: Kokoro / Chatterbox Turbo Music Generation: ACE-Step-v1.5-XL-Turbo for rendering session summaries into lyrics/balladsA Few Engineering Lessons & Challenges:* State Delta Extraction vs. Context Explosions: Feeding 20 prior session transcripts into context windows quickly becomes cost-prohibi
Progress Software Announces New Telerik and Kendo UI Release to Accelerate AI-Powered UI DevelopmentBy embedding AI deeply into UI generation, application modernization and development workflows, the release enables teams to move beyond traditional software development cycles to AI-assisted coding, ...
Why Finance Teams Aren’t Using The Tools CFOs Bought ThemFinance is using AI, but rarely at scale. The real barrier isn’t technology. It’s getting people to trust, adopt, and embed AI in daily work.
Show HN: Lutefish Stream (Play Music over Internet)Hey all, Disclaimer: I work on embedded software for Lutefish.tldr: If you have good internet and you live within 500 miles of somebody (and ISPs are feeling benevolent that day) you can play music together over the internet (for real tho).I know you've probably seen "make music over the internet but for real this time" presented from a handful of places over the years. We designed a PCB from scratch to tightly couple the audio and networking layers as close as we possibly could, and we are able to get latency numbers as low as 28ms using custom protocol (which allows for seemingly real-time performance with bandmates over the internet).Before you ask, no, it is not an audio interface also (we're working on it). Think of it more like, "point-to-point sound cards with a lot of control over your mix".Don't really know how to do self promotion, but we have devices for sale for $250 usually (rando $200 flash sales sometimes) and there's a $10 / month subscription (2 months free w/ purchase
MIC Global and inDrive Expand Insurance Partnership to 18 Countries, Extending Protection to Millions of Drivers and Passengers WorldwideMIC Global is built for exactly this: short duration, event-driven, automated claims and embedded insurance. Its triggers are binary and platform-verified, so claims are adjudicated using data the ...
Show HN: Minna – On-device semantic search across your own documents (macOS)Hi HN! Minna is a new tool geared towards people who have a knowledge base so big they forget where things are.- Chat answers are constrained to retrieved results. The model has to search your knowledge base first and cite the document it pulled from. It won't answer from outside your knowledge base which means you can trust the LLM instead of having to be weary about what it outputs. - Hybrid retrieval — BM25 over an inverted index for exact and keyword matches, plus a vector index for semantic matches, results are fused using Reciprocal Ranked Fusion. - Embeddings and inference both run on-device. Search uses bge-small-en and run on the neural engine. Chat uses Qwen3 4B running on MLX. - Search latency is ~12 ms across 250 documents split into 30,000 chunks on Apple Silicon.What’s not done; OCR scanning on PDFs is limited, the app is currently Apple Silicon only, and there are some rough edges in the UI.It’s a TestFlight beta (apologies, but it is the cleanest distribution path right
Show HN: Superprez.io – Sharing and Collaboration for AI-Generated DecksHi HN, I've been working on Superprez.io because of something that started bothering me more and more with AI-generated presentations.Any AI like Claude, Cursor and ChatGPT can now generate surprisingly good decks — not just visually, but with animations, interactive elements, embedded products, maps, live polling, etc.But a lot of these presentations are ultimately treated like traditional presentations: export them to PDF or PPT and send them around. And that's where a lot of what makes them interesting disappears.If the presentation is actually built with HTML, CSS and JS, why turn it back into a static file?That's what we're trying to solve with Superprez.io: a way to share and collaborate on presentations as code, through a simple URL, without the constraints of the traditional tools.The interesting part for me is that AI-generated presentations make this problem much more obvious. Once you're generating presentations as code, there are a lot of things you can do that simply don't
Chinese automotive glass giant announces vehicle-integrated solar roofFuyao Glass has confirmed mass-production capabilities for solar sunroof glass, with photovoltaic cells embedded in laminated automotive glass to power auxiliary vehicle loads. While reports have ...
Show HN: UniFi OS Server in DockerUbiquiti only ships the UniFi OS Server (UOS) as an OCI image embedded inside a proprietary Rust installer — you can't just pull it from a registry. This repository extracts that OCI image, adjusts it to our liking, and ships it with a compose file and config that make it fit to run in Docker. No parallel Podman setup required.
Show HN: Visimer – open-source visual editor for Mermaid diagramsHi HN, Nick here, founder at OpenKnowledge. We built an open source library, Visimer, for editing Mermaid diagrams with a what-you-see-is-what-you-get visual editor. Try the playground at https://visimer.com/playground. You can rename labels, change shapes, add new nodes, etc. by clicking and dragging the visual canvas.We set out to build this originally for OpenKnowledge, our markdown wysiwyg editor, then realized it’d be great to share it as a proper embeddable component library so that other apps can leverage it as well.Our general take is that AI is great for generating content, but whenever you want to tweak fine details, for example to create an artifact sharable with others, point-edit functionality is much better. We built Visimer to bring that to Mermaid diagrams, just like we did for markdown.Some technical bits:- Leverages the native Mermaid.js renderer for visualizing the Mermaid diagram true to how it’s intended. We add point-edit functionality on top.- We tried re-generat
Google Lets You Turn Off Visible Watermarks In GeminiGoogle Gemini allows users to disable the visible watermark on AI-generated images, videos, and music, while SynthID and C2PA marks stay embedded.
Show HN: Burla – Distributed computing framework for AI agentsHi HN! Would love any thoughts on this, we built Burla to enable anyone, even total beginners to scale Python to thousands of computers in their cloud with zero hassle. In a world of coding agents, this means something different than it used to, specifically that Burla requires almost no cloud permissions to get started. Anyone who has permission to boot a VM in their cloud can simply pip install burla and scale Python to 1000's of VM's. Burla uses your local aws cli credentials to boot VM's and the dashboard runs locally for you to monitor resources and jobs.All together this creates a UX where you can simply ask your agent to do something like "Vector embed all of Wikipedia" and it will do it far faster and with less compute than it would have using any cloud service.
Show HN: I spent almost 2 years collecting website design inspirationNone of the captures are scraped data it's all painstakingly manual albeit crowd-sourced work.You can see per-section captures for a given website with their corresponding colors and gradients, typography information (weights, sizes, line heights, letter spacing, etc), border styles and whole lot more. Under font family information i also capture foundry, designer related information parsed from the file's metadata.For similarity search it stores visual embeddings for semantic search and glyph similarity for font similarity searches.There's also a well-defined taxonomy for more coarsely grained searches.
The 10 Best Pet Hair Vacuums Designed for Homes That Never Stop SheddingGetting pet hair out of carpets and off of tile and hardwood is a constant battle many are familiar with, and trying to remove those persistent hairs that have embedded themselves into your upholstery ...