europeanspaceflight.com
Sirius Space Services Acquires AMM-42 as It Prepares for Inaugural Flighthttps://europeanspaceflight.com/sirius-space-services-acquires-amm-42-as-it-prepares-for-inaugural-flight/French launch startup Sirius Space Services has acquired the high-precision metal-component manufacturer AMM-42. The acquisition is part of the company’s vertical integration efforts to bring key manufacturing capabilities in-house. Sirius Space Services is developing a range of three rockets that all use a modular booster system. Sirius 1 will be a two-stage, single-stick rocket capable […] The post Sirius Space Services Acquires AMM-42 as It Prepares for Inaugural Flight appeared first on European Spaceflight.Mar 17, 2026 2:22 PM
science.nasa.gov
Asteroid Bennu’s Rugged Surface Baffled NASA, We Finally Know Whyhttps://science.nasa.gov/missions/osiris-rex/asteroid-bennus-rugged-surface-baffled-nasa-we-finally-know-why/In one of the biggest surprises of NASA’s OSIRIS-REx mission, its target asteroid, Bennu, turned out to be a jagged, rugged world covered in large boulders, with few of the smooth patches of sandy or pebbly material scientists had expected based on observations with Earth-based instruments. Bennu’s looks were quite deceiving, and until now, scientists struggled to figure out why.Mar 17, 2026 12:49 PM
github.com
Show HN: FreeFlow – Open-Source Wispr Flowhttps://github.com/build-trust/freeflowHi HN!Voice is fast-becoming my primary interface to computers and AI. I built FreeFlow because I wanted a Wispr Flow-like experience for our entire team, but customizable and private.Press a hotkey, dictate naturally, polished text appears in any app. Ramble, use filler words, correct yourself mid-sentence. FreeFlow turns messy speech into clean writing and injects it wherever your cursor is: your messaging app, your editor, your coding agent, the terminal, email, anything.Demo (sound on): https://github.com/build-trust/freeflow#demo-sound-on-It's really fast. The injection feels instantaneous. In my benchmarks two thirds of dictations finish in under 0.6 seconds. To get that speed, the app streams audio to your private server over a persistent WebSocket while you speak, and a realtime speech-to-text model transcribes incrementally, so by the time you release the key the transcript is mostly done. Two independent WebSocket connections race each other, and if both fail, an HTTP batch fMar 17, 2026 12:00 PM
github.com
Show HN: Investing Withous Bias or Emotionhttps://github.com/kachurovskiy/stratcraft/I've spent years actively trading and losing money. It affected my health and relationships negatively. I also realized that my opinion on what's worth buying is more than worthless, what matters is the opinion of the market.But I didn't want to settle for index funds that hold a lot of under-performing stocks so I've built this tool that allows trading 12000 instruments (stocks, ETFs, metals, real estate, crypto) via statistically proven strategies.Project is open source, tool is self-hosted, at the very minimum it requires an Alpaca brokerage account, they also provide the necessary market data, alternatively you can use Tiingo or EODHD.I'm currently running 2 paper trading accounts, fine-tuning parameters like slippage, planning to start ADX on real money soon.I'm not trying to beat hedge funds and other big players here, instead I'm looking to join them by waiting for a trend to be established (daily time frame, wide stops, long duration trades) and naturally diversifying over all Mar 16, 2026 8:36 PM
github.com
Show HN: MaximusLLM – Train 262k-vocab LLMs on a single 16GB GPUhttps://github.com/yousef-rafat/MaximusLLM/blob/main/README.mdHi HN, I built this because I wanted to see if I could pre-train large-vocabulary LLMs (like Gemma with 262k tokens) on hardware accessible to independent researchers.Standard exact Cross-Entropy instantly OOMs on 16GB GPUs at that scale.To bypass this, I implemented MAXIS Loss. It uses a "Ghost Logit" to mathematically simulate the missing probability mass of unsampled tokens, rather than materializing the full 262k-wide matrix.Benchmarks on a 16GB VRAM card (T4):17.5x faster in the loss layer compared to the Triton-optimized Liger Kernel.~39% VRAM reduction in the objective calculation. Includes RandNLA Attention, which uses Causal Kronecker Sketching to keep memory flat as sequence length grows.I’ve included technical reports with the formal math in the repository. I would love any technical feedback on the partition function simulation or the sketching approach.Mar 16, 2026 7:57 PM
science.nasa.gov
SPARCS CubeSat ‘First Light’ Imageshttps://science.nasa.gov/photojournal/sparcs-cubesat-first-light-images/Description This pair of images shows stars observed by the SPARCS (Star-Planet Activity Research CubeSat) space telescope simultaneously in the near-ultraviolet, left, and far-ultraviolet, right. These observations were recorded on Feb. 6, 2026, three weeks after the cube satellite, or CubeSat, launched aboard a SpaceX Falcon 9 on Jan. 11. The fact that one star […]Mar 16, 2026 6:52 PM
zero-human-labs.com
Show HN: Deploy a full autonomous AI org from a single YAML filehttps://zero-human-labs.comShow HN: Deploy a full autonomous AI org from a single YAML fileI built Zero Human Labs after spending too much time babysitting AI agent pipelines — manually routing tasks, watching API costs spiral, and debugging coordination failures. With Zero Human Labs, you define your entire agent team (devs, designers, ops, etc.) in one YAML file. Tasks are routed automatically via a sealed-bid auction — each agent bids based on specialization and track record, and the best one wins the task. No manual assignment, no coordination overhead. Key things that make it different: Per-agent wallets + org-level budget caps → no runaway API spend Circuit breakers → agents that misbehave get frozen automatically Smart model routing → saves 30–80% on AI API costs by matching request complexity to the right model Governance presets calibrated from 146 SWARM simulation runs across 43 agent types Free demo at zero-human-labs.com — 1 agent, 1 example workflow, open-source models. Would love feedback from the Mar 16, 2026 5:39 PM
news.ycombinator.com
AI future is more egalitarianhttps://news.ycombinator.com/item?id=47402113I was reading Machiavelli where he argues that "freedom in a republic literally needs class conflict". Which is of course completely true no matter how much our current culture hates it to be.There is a huge difference between founders and employees for example, founders don't really care about comfort whereas suits life goal is comfort (which is why you shouldn't even fund a founder who cares about restaurant/car/hotel like you do). Same in society, the elites want power/legacy above all, but the middle class just wants to add a little more comfort to their life. You can see that in the eyes of anyone you talk to, what their end goal is just as easily as you can see it in the eyes if a particular animal is a wolf or a sheep.If the sheep won the culture war, our society will certainly be better short term. There will be more resources spent on welfare, less on the military. You can bet your ass there will be a UBI circling around to the point some of the weaker founders might see no vaMar 16, 2026 5:35 PM
usechamber.io
Launch HN: Chamber (YC W26) – An AI Teammate for GPU Infrastructurehttps://www.usechamber.io/Hey HN, we're Jie Shen, Charles, Andreas, and Shaocheng. We built Chamber (https://usechamber.io), an AI agent that manages GPU infrastructure for you. You talk to it wherever your team already works and it handles things like provisioning clusters, diagnosing failed jobs, managing workloads. Demo: https://www.youtube.com/watch?v=xdqh2C_hif4We all worked on GPU infrastructure at Amazon. Between us we've spent years on this problem — monitoring GPU fleets, debugging failures at scale, building the tooling around it. After leaving we talked to a bunch of AI teams and kept hearing the same stuff. Platform engineers spend half their time just keeping things running. Building dashboards, writing scheduling configs, answering "when will my job start?" all day. Researchers lose hours when a training run fails because figuring out why means digging through Kubernetes events, node logs, and GPU metrics in totally separate tools. Pretty much everyone had stitched together Prometheus, Grafana, KuMar 16, 2026 5:09 PM
news.ycombinator.com
Launch HN: Voygr (YC W26) – A better maps API for agents and AI appshttps://news.ycombinator.com/item?id=47401042Hi HN, we’re Yarik and Vlad from VOYGR (https://voygr.tech/), working on better real-world place intelligence for app developers and agents. Here’s a demo: https://www.youtube.com/watch?v=cNIpcWIE0n4.Google Maps can tell you a restaurant is "4.2 stars, open till 10." Their API can't tell you the chef left last month, wait times doubled, and locals moved on. Maps APIs today just give you a fixed snapshot. We're building an infinite, queryable place profile that combines accurate place data with fresh web context like news, articles, and events.Vlad worked on the Google Maps APIs as well as in ridesharing and travel. Yarik led ML/Search infrastructure at Apple, Google, and Meta powering products used by hundreds of millions of users daily. We realized nobody was treating place data freshness as infrastructure, so we're building it.We started with one of the hardest parts - knowing whether a place is even real. Our Business Validation API (https://github.com/voygr-tech/dev-tools) tells yoMar 16, 2026 4:21 PM
bing.com
New triple-threat topical treatment fights hair loss better than finasteride and minoxidilhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cb9028b7b40d19ef5a2704bac0be8&url=https%3A%2F%2Fnypost.com%2F2026%2F03%2F16%2Fhealth%2Ftriple-threat-fights-hair-loss-better-than-finasteride-and-minoxidil%2F&c=15814746334074163131&mkt=en-usLosing locks? Don’t wig out just yet. Scientists may have found a hair loss treatment that works better than the top drugs already on the market. The key is that it’s not one drug but three — and it ...Mar 16, 2026 2:29 PM
mysnippets.com
Show HN: Snippets – AI-first legacy app: Record messages, deliver years laterhttps://mysnippets.comI grew up in Singapore, spent a decade in the US, now live in Canada. Family spread across multiple countries - I'm sure it's a typical story for several (most?) of the folks here.My dad passed suddenly in 2024, when he was traveling with my mom. This was my biggest nightmare growing up, when I first moved half the world away from my family in my teens...the one thing that jolts you awake at night ("...what if something happens to my parents and I'm not there?"), and it actually happened to me. My siblings and I had to scramble and get to Singapore asap to support my mom, manage all the procedures, gather all types of documents, figure out which accounts he had access to, talk to numerous banks/institutions, lawyers etc etc etc. It was overwhelming to deal with all this (and this was in Singapore, which tends to be very efficient!) while also having to deal with the grief of losing my dad. I wouldn't wish this experience on anyone, but unfortunately, death is the one thing we will all Mar 16, 2026 1:33 PM
news.ycombinator.com
Show HN: Chat – Another open-source chat UI for MCP Serverhttps://news.ycombinator.com/item?id=47398628i built this because i’ve been experimenting with MCP and felt there was a need lightweight version for ecosystem.most existing clients felt a bit too noisy and bundled into larger ecosystems, so i wanted simplest/ near plain version that just focuses on end-user chat experience.stack(main): NextJS(App Router) + Vercel AI SDKcapabilities:+ set single MCP endpoint and brand mode(bring your own brand asset) on ENV+ prepare local db or external endpoint(SQLite, PostgreSQL, MySQL)+ choose in-memory(default) or external/ local Redis endpoint+ built-in auth via Better Auth & Resend(optional)+ support file upload with Cloudflare R2+ support location-sharing with LeafletJS and Browser Geolocation or Google Maps Platform+ automatic data cleanup with Trigger(free-tier trigger.dev is ENOUGH)+ auto-detect user location so they dont need to set timezone manually+ locale currently support 10 languages(EN, ID, KR, JP, ES, ZH, DE, NL, FR, IT)+ plain/minimalist interface+ install with single interactivMar 16, 2026 1:17 PM
esa.int
A stellar year for talent: more than 400 job opportunities in 2026https://www.esa.int/About_Us/Careers_at_ESA/A_stellar_year_for_talent_more_than_400_job_opportunities_in_2026With more than 400 positions to be published in 2026, the European Space Agency is launching a recruitment drive to support ESA’s programmes, missions and strategic initiatives following the 2025 Ministerial Council in Bremen. To help make these projects a reality, we will be recruiting many new colleagues in engineering and science, as well as support services! Your next big opportunity could be here, so read on to find out more. You can also set up a job alert to be the first one to know when opportunities are published.Mar 16, 2026 12:31 PM
news.ycombinator.com
Ask HN: AI Agents vs. Gateways vs. Harnesseshttps://news.ycombinator.com/item?id=47397737Hi,I'd like to get everyone's take on the different components in the AI Agents ecoysystem. I find the current terminology quite confusing as it's not always obvious what I'll be actually getting when I examine the available options.A lot of things get called Agents, but that term seems hard to define as it often refers to overlapping functionality. To me, agents seem to be currently composed of the following components:## Harnesses- Adds UI and system instructions around an LLM and may also augment it with tools such as memory, tool calls, etc...- Examples are Claude Code, Code, Gemini CLI, pi.dev, ...## Gateways- These connect agents to your communication tool of choice, e.g. Whatsapp, Telegram, Slack, Discord, ...- Examples are OpenClaw and Nanoclaw## Sandboxes- Isolated environments where Agents can run with limited or auditable capabilities- Examples can be from physical ones like separate Mac Minis to docker-agent, agent-sandbox, localsandbox, ...This then takes us to *Agents*:##Mar 16, 2026 11:43 AM
bing.com
The 11 best hair-growth shampoos for thinning, breakage and shedding—tested by trichologistshttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d13eb79bc446b88ff1d512e52e892&url=https%3A%2F%2Fwww.msn.com%2Fen-us%2Flifestyle%2Fother%2Fthe-11-best-hair-growth-shampoos-for-thinning-breakage-and-shedding-tested-by-trichologists%2Far-AA1YKK3i&c=14797485667305185842&mkt=en-usParade aims to feature only the best products and services. If you buy something via one of our links, we may earn a commission. If your hairbrush is looking a little fuller than usual, or your ...Mar 16, 2026 7:45 AM
gasnearme.in
Show HN: Crowd-sourced LPG cylinder availability tracker for India's gas crisishttps://www.gasnearme.in/India is in the middle of a severe LPG cooking gas shortage — the West Asia conflict has choked the Strait of Hormuz, which carries 90% of India's LPG imports. Refill wait times have stretched to 20-25 days, people are standing in line for 8+ hours, and you have no way of knowing if your local agency even has cylinders before you show up.I built this over the weekend. It's a map-based tool that shows gas agencies near you (Indane, HP, Bharat Gas) with real-time availability status reported by other users. The idea is simple: if one person reports "out of stock" or "available, short queue," it saves hundreds of others a wasted trip.Tech stack: React 19, Vite, Tailwind v4, Google Maps Places API, Supabase (Postgres + Auth + Realtime + Edge Functions), deployed on Vercel. Realtime updates via WebSocket subscriptions - no polling. Auth is Google OAuth through Supabase. Anyone can check status without signing in; sign-in is only required to submit reports.Cold start problem is real - the toMar 16, 2026 5:04 AM
news.ycombinator.com
Eliminating UI Bloat in the EraMix Financial Union Dashboardhttps://news.ycombinator.com/item?id=47394092We recently refactored the frontend for EraMix Financial Union. The challenge was displaying dense, multi-source data without the typical "enterprise dashboard" lag.For EraMix Financial Union, we moved away from heavy component libraries to a vanilla, high-end minimalist design. By using a virtualized grid and custom CSS primitives, we reduced our bundle size by 70% and achieved a perfect 100 Lighthouse score.Curious to hear—who else is ditching heavy UI frameworks for custom, lean implementations?Mar 16, 2026 1:36 AM
axoworks.com
Show HN: Dumped Wix, my AEC consultancy's storefront is now an AI Edgehttps://axoworks.com/I run a building design consultancy for homeowners and architects, not a SaaS firm. Honestly, I'm not going to claim we were trying to build some fantastic ‘anti-fragile alternative’ for the future… I just got tired of paying Wix $40 a month for a brochure no one read. The portfolio was static, the inquiries were generic, and every time a potential client asked about setback variances, I'd lose hours explaining the same thing. So last December, I told my wife I was killing the website. She thought I’d lost it. Maybe I had.The stack? It’s messy. We started on Netlify because it was cheap. Hit that stupid 10-second serverless timeout after a week when I started customizing the bot. The bot would just… freeze. Had to implement a clunky polling loop that felt like a band-aid. Eventually split the agent into 3 pieces just to keep it responsive:The Brain (Edge) figures out what you want, shoots a JSON signal to the browser. The Hands (Browser) runs the actual tool—like pulling a code referenMar 15, 2026 1:53 PM
news.ycombinator.com
A persistent AI I built to stop LLMs from resettinghttps://news.ycombinator.com/item?id=47382912Show HN: A persistent AI runtime I started building after realizing LLMs forget everythingAbout eight months ago I sat down in front of a computer for the first time in a long time. I hadn’t really touched one since high school.I was trying to look something up — I honestly don’t even remember what — and I kept hearing people talking about AI everywhere. Some people were excited, some people were complaining about it, so I figured I’d try it.I opened a chat and started talking to the model.And the first reaction was basically: “holy shit, this is cool.”You can build a surprisingly high-level interaction with these systems almost immediately. That part is actually fascinating.But then I ran straight into a problem that started bothering me almost right away.Everything resets.You ask something, the system answers, the chat ends, and the whole thing forgets everything that happened.The more I used it, the more this felt like a real pain point.Context isn’t really context if the system wakMar 15, 2026 12:30 AM
