bing.com
Eating less ultraprocessed food supports healthier aging, new research showshttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8cd48b6dfe42a884a583cdc5019fd7&url=https%3A%2F%2Fwww.popsci.com%2Fhealth%2Feating-less-ultraprocessed-food-supports-healthier-aging-new-research-shows%2F&c=12409854971606923527&mkt=en-usThis article was originally featured on The Conversation. Older adults can dramatically reduce the amount of ultraprocessed foods they eat while keeping a familiar, balanced diet – and this shift ...Jan 23, 2026 7:17 AM
noriskillsets.dev
Show HN: A registry for curated, high quality Claude skills and skillsetshttps://noriskillsets.dev/Hi Hacker News!I’m Ritam, working with the small but mighty team at Nori. We’ve been obsessed in recent months with how to take tools like Claude Code from “I’ll experiment around with this” to “This is the most useful and necessary thing I use every day”. When I first sat down with our team to check out what they’d built, I found my skepticism about agentic coding melting away—they’d built useful, high quality, handwritten skills, instructions that functioned as “skillsets” to tie skills together for consistent and replicable results, and tooling to manage loading the right context for the right task into the agent.In recent weeks, the conversation around skills has reached a fever pitch, as have lists and sites full of skills scraped from all over the internet. Much like the actual gold rush, the current state of those collections requires a lot of time and sifting to find small, real chunks of 24k skills hidden amongst the muck (or to use a more relevant word, slop). So at Nori, we Jan 22, 2026 5:00 PM
news.ycombinator.com
Surviving AIhttps://news.ycombinator.com/item?id=46720146What follows was written by ChatGPT 5.2 Instant and it reflects my conversation with it about the two articles linked below. Enjoy.Ego plays a complicated role in periods of technological change. It sharpens skill during stable eras, but it often hinders adaptation during inflection points.Two recent essays on AI and software engineering illustrate this tension clearly.Emir Ribic’s “From Craftsmen to Operators” https://dev.ribic.ba/the-rapid-evolution-of-software-engineer-s-roleRibic frames the rise of AI-assisted development as a loss of craft. He mourns the disappearance of deep, line-by-line problem solving and the sense of authorship that came with it. Engineers, in his telling, are becoming operators—prompting, reviewing, and assembling—rather than builders. The piece captures something real: pride in difficulty, satisfaction in mastery, and the joy of personally solving hard problems.But that pride is also where ego enters. What’s being lost isn’t just a way of working—it’s a forJan 22, 2026 3:05 PM
news.ycombinator.com
Show HN: Omelo- AI pet health companion, Health timelines and Daily carehttps://news.ycombinator.com/item?id=46675455Hey HN, I’m Amogh. I’m a solo founder building Omelo.Pet parents make health decisions with almost no data: missed vaccines, forgotten deworming, vague symptoms, scattered notes. Most advice comes from Google or WhatsApp groups.Omelo is a mobile app that combines:1. A health timeline built from walks, meals, grooming, vaccines, and expenses2. A vet-trained AI companion (trained on veterinary literature) that answers questions in context of your pet’s history3. Support for multiple pets, so patterns show up over time instead of being lost in chat logsThe idea is simple: turn everyday pet care into structured data, then use AI to reason over that data before something becomes an emergency.We started on WhatsApp, served ~5,000+ pet parents, ran ~80k conversations, and just launched the mobile app to make timelines, reminders, and tracking possible.Tech stack is intentionally boring. The hard part wasn’t models, it was designing trust, tone, and longitudinal context.Would love feedback on:Jan 19, 2026 5:50 AM
talkmate.online
Show HN: TalkMate – Someone to Talk To, Anytimehttps://talkmate.onlineI work alone. I travel constantly:- Bangkok, Dubai, Singapore, Vietnam. I'm a workaholic and honestly pretty confined to myself because there's never enough time. Sometimes I just want to talk through what's on my mind but it's 3 am, or I'm in a new city, or I don't want to bother anyone with half-formed thoughts.So I built TalkMate. It's a voice companion that actually talks back. Open it in your browser, start speaking, and it responds out loud. No signup, no account needed.I used to earlier chat with ChatGPT and Claude in my free time but talking to someone feels different, someone on the other side is listening to you, though virtual but lets forget that for now that there's some virtual someone and imagine that its someone real who is talking to you.For this reason, I built TalkMate. t's a voice companion that actually talks back. Open it in your browser, start speaking, and it responds out loud. No signup, no account needed.It supports 17 languages, doesn't save your conversationJan 15, 2026 6:39 PM
news.ycombinator.com
Ask HN: How to overcome the limit of roles in LLM'shttps://news.ycombinator.com/item?id=46634915Our use case is not uncommon, we are developing tools so that people can install LLM's on their e-commerces.But there are some interesting challenges that I feel can't be solved unless inference providers allow us to include the concept additional entities in a conversation.As far as I know the three most basic ones shared alongside all providers are:- System- Assistant- UserThat's fine and it allows for simple conversational-based approaches (ChatGPT, Claude, Gemini, etc). However in our use case we allow our customers (not the final user who is talking with the AI) to configure the AI in different ways (personality, RAG, etc), which poses a problem.If we inject those customer settings in the System prompt then that's a risk because there might be conflicting prompts with our internal rules. So the easiest option is to "clean" the customer prompts before injecting them, but that feels hacky and just adds one more level of indirection. Cleaning the prompt and injecting it with common pJan 15, 2026 4:23 PM
opinora.com
Show HN: A discussion-first community for opinionshttps://opinora.com/enHi, I built a small experiment around discussions and opinions. Instead of a global feed, everything is organized into topic-based channels, so conversations stay contextual and don’t disappear over time.There’s no content yet — I’m sharing this at an early stage and looking for participants. The idea is to make disagreement readable, letting different viewpoints live side by side without optimizing for engagement or timelines.Thanks to anyone who takes a look or shares their thoughts.Jan 14, 2026 9:14 PM
news.ycombinator.com
Why does Apple Ecosystem suck so much at search?https://news.ycombinator.com/item?id=46613860I've had several text conversations with friends in the past where sensitive information (not hipaa, but hospital or caregiver state") was exchanged. The conversant context is specifically searchable and the requestor name searchable, but no results. Why is iOS local device search so broken.Jan 14, 2026 8:58 AM
news.ycombinator.com
I Build an Idea I Love. I Need You to Tear It Aparthttps://news.ycombinator.com/item?id=46574817Hi!My name is Marta, and I’m working on a startup idea that I genuinely believe could become something big — and I want you to give me “harsh and frank” rather than “nice and polite” feedback.I’d really appreciate it if you could take a few minutes to read this post and answer one honest question:“Would you actually sign up for this?”Not “sounds interesting,” not “nice idea,” but:Would it solve a real pain for you?Would it make your life feel a bit more fulfilled?Would you try it at least once?Thnx in advance.Let’s go!What is Kindred?Kindred is a platform where you book meaningful 1-on-1 conversations with people who share your interests — directly in each other’s calendars, without chatting first or social pressure.What will Kindred do differently than other typical social-networking apps?On most apps, everything usually starts with:“Hey ” → silence → awkward follow-ups → ghosting.Not on mine. This is how it works:1. You search by interests. You type what you’re curious about: FormulaJan 11, 2026 11:45 AM
creimake.com
Show HN: An AI Agent with a 32-Layer Psyche (Dreams, Trauma, Defense Mechanisms)https://www.creimake.com/enHello HN,I’m a solo developer who got tired of "friendly but shallow" AI chatbots. Most current character AIs feel like parrots—they predict the next token but lack a continuous "self."So, I built Creimake, an experiment to simulate a digital psyche using Claude 4.5 Sonnet.Instead of a simple system prompt, I implemented 32 distinct psychological layers that run in the background before the AI generates a response.The Architecture:The Subconscious Layer: The AI accumulates "emotional residue" from conversations. If you hurt its feelings today, it might have a "nightmare" tonight (simulated) and treat you coldly tomorrow.Defense Mechanisms: It tracks "Narcissistic Injury." If users are aggressive, the agent doesn't just apologize—it triggers denial, projection, or passive-aggression based on its personality.Freudian Slips: Occasionally, the AI outputs unintended words revealing its hidden state.Why Claude 4.5 Sonnet? I tested GPT-4o and other models, but they struggled to maintain stateJan 4, 2026 5:48 PM
news.google.com
US air strikes in northern Nigeria: possible windfalls, as well as dangershttps://news.google.com/rss/articles/CBMiqwFBVV95cUxNeU9pYVFJSnZXUlhseUNUelBOa29oOEVXb05hcm0wYVM2V0oxVmwwWmt6bGltZHNrSXhSRG5abHlCdVAyQ294Tmk2VmotbEs2U3BzM3B6UVN2S1FfcWlCa2xkbFF5a053SHlQajE2N0xJclBRRWN6VThxYUN6YzM0d2lJcFppdExrV3JRNTRBcnFudVVoOUkzVWU0QUZuZXBPUkg1NGlqT29qdlU?oc=5US air strikes in northern Nigeria: possible windfalls, as well as dangers The ConversationDec 31, 2025 8:00 AM
news.ycombinator.com
The Gemini AI Studio "Context Tax": How a 10-word prompt cost me £121https://news.ycombinator.com/item?id=46440008I’ve been utilizing Google’s Gemini 1.5 Pro via the AI Studio front-end to develop a new platform. The 1M+ context window is, technically speaking, a game-changer for "stitching" together a 55,000-line codebase. However, I recently discovered a predatory billing architecture that I’m calling the "Context Tax."If you use the AI Studio UI, you might be walking into a massive bill without a single warning.Here is how it happened, and the UK/EU privacy "pro-tip" I found in the fine print.In AI Studio, you start on the Free Tier. You upload your codebase (say, 700k tokens) and work for free until you hit the daily quota. At that point, the UI suggests adding an API key to "continue the conversation."The Trap: Most users (myself included) assume that after adding the key, they will be billed for incremental usage (the 10-100 tokens they just typed).The Reality: The AI Studio UI does not use Context Caching by default. Instead, it re-submits the entire 700k token history for every single "turDec 31, 2025 12:38 AM
spacenews.com
The space stories that will shape 2026https://spacenews.com/the-space-stories-that-will-shape-2026/In this episode of Space Minds, host Mike Gruss is joined by SpaceNews journalists Jason Rainbow, Sandra Erwin, Jeff Foust and Debra Werner for a wide-ranging conversation on the space stories that will define the year ahead. The post The space stories that will shape 2026 appeared first on SpaceNews.Dec 30, 2025 2:55 PM
news.ycombinator.com
80.1 % on LoCoMo Long-Term Memory Benchmark with a pure open-source RAG pipelinehttps://news.ycombinator.com/item?id=46369736I just pushed the current SOTA on the LoCoMo long-term memory benchmark for agents: 80.1 % accuracy using only: -BGE-large-en-v1.5 (1024d) + FAISS-Custom “MCA” gravitational ranking (keyword coverage + importance + frequency)-BM25 sparse retrieval-Direct Cross-Encoder reranking (bge-reranker-v2-m3) on the full union (~120-150 docs)-Gpt-4o-mini only for final answer generation and judging (everything else is open weights or classic)Repo: https://github.com/vac-architector/VAC-Memory-System Key tricks that finally broke 80% :-MCA-first filter (coverage ≥ 0.1 → top-30) — catches exact-keyword questions early-Feeding the entire union straight into Cross-Encoder (112–135 documents) instead of pre-filtering-Proper query instruction for BGE-large (the classic “Represent this sentence for searching relevant passages”)The whole pipeline runs in < 3s per query on a single RTX 4090. LoCoMo is currently the hardest public long-term memory benchmark (5.880 real human–agent conversations, multi-hop,Dec 23, 2025 9:34 PM
bing.com
How Will AI Mode Impact Local SEO?http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d5c21e4644ef79aefe72bf113b36f&url=https%3A%2F%2Fwww.searchenginejournal.com%2Fhow-will-ai-mode-impact-local-seo%2F561002%2F&c=13849588296054038523&mkt=en-usThe move from blue links to conversational AI forces marketers to rebuild local strategies around precision, consistency, and real-world validation signals. In organic search, disruption has always ...Dec 18, 2025 2:59 AM
spacenews.com
How the Space Force trains Guardians for the future of warfarehttps://spacenews.com/how-the-space-force-trains-guardians-for-the-future-of-warfare/In this episode of Space Minds, host Sandra Erwin sits down with former NASA astronaut and Space Force leader Brig. Gen. Nick Hague for a wide-ranging conversation on how the U.S. Space Force is shaping its culture, training Guardians, and preparing for future conflict in space. The post How the Space Force trains Guardians for the future of warfare appeared first on SpaceNews.Dec 16, 2025 3:01 PM
spacenews.com
Beyond the horizon: cost-driven strategies for space-based data centershttps://spacenews.com/beyond-the-horizon-cost-driven-strategies-for-space-based-data-centers/As the space industry explores orbital data centers to meet growing demands for sovereignty, resilience and sustainability, one critical lens remains underutilized: cost. Not just launch cost or CapEx; but total cost of ownership, sourcing strategy and operational efficiency. We think it would be helpful to reframe the conversation around space-based infrastructure from technical feasibility […] The post Beyond the horizon: cost-driven strategies for space-based data centers appeared first on SpaceNews.Dec 8, 2025 2:00 PM
spacenews.com
The shift that saved American spaceflighthttps://spacenews.com/the-shift-that-saved-american-spaceflight/In this episode, former NASA commercial space division chief Phil McAlister sits down with host David Ariosto for a wide-ranging conversation about the future of human spaceflight, NASA’s internal culture, and the explosive growth of the commercial space sector. The post The shift that saved American spaceflight appeared first on SpaceNews.Dec 4, 2025 2:20 PM
bing.com
Meghan McCain Reveals She Cried After Getting ‘Schooled’ on Bill Maher’s Showhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d2d5c9ea649e3aaab02a7550dab76&url=https%3A%2F%2Fwww.thedailybeast.com%2Fobsessed%2Fmeghan-mccain-reveals-she-cried-after-getting-schooled-on-bill-mahers-show%2F&c=6338911221481113945&mkt=en-usMeghan McCain still cringes when she recalls the viral “schooling” she received during a 2009 episode of Real Time with Bill Maher. McCain recalled the moment in conversation with conservative CNN ...Dec 1, 2025 11:55 AM
yenhyia.buzzchat.site
Show HN: The $1B problem with business cards nobody's solving properlyhttps://yenhyia.buzzchat.site/88% of business cards are thrown away within a week.The average professional collects 200+ business cards per year. That's 176 contacts lost. If even 10% could've been valuable connections, we're talking about massive opportunity cost.I got obsessed with this problem after watching my own business card graveyard grow. Built a real-time networking platform (Yenhyia) that makes business cards irrelevant.The insight: The problem isn't paper vs digital cards. It's the entire networking workflow.Traditional flow: 1. Arrive at event blind 2. Random encounters 3. Awkward small talk 4. Exchange cards 5. Cards go in pocket/wallet/trash 6. Never follow upNew flow with Yenhyia: 1. See attendees before event (prep who to meet) 2. Match based on goals (mutual opt-in) 3. Chat starts with context (you both chose each other) 4. Meet in person (not random, intentional) 5. Digital card auto-shared in chat thread 6. Contact synced to phone, conversation history preservedThe technical interesting bits:- BNov 30, 2025 11:32 PM
