spacedaily.com
UNSW Canberra team advances iodine propulsion for satelliteshttps://www.spacedaily.com/reports/UNSW_Canberra_team_advances_iodine_propulsion_for_satellites_999.htmlSydney, Australia (SPX) Feb 18, 2026 Thousands of satellites are launched every year to support navigation, telecommunications, weather forecasting and emergency warning services, and most rely on onboard propulsion systems to manoeuvre once in orbit. These propulsion systems are essential to change orbits, perform end of life de-orbiting, carry out collision avoidance with space debris or other satellites, maintain or adjustFeb 23, 2026 2:45 PM
news.ycombinator.com
Show HN: I built an async Telegram bot to snipe 80%+ retail price errorshttps://news.ycombinator.com/item?id=47116609Hi HN,I do some retail arbitrage and got incredibly frustrated missing out on massive 90% off glitches. By the time someone posts a deal on Slickdeals or DealNews, bots have already cleared the stock.I decided to fix my own problem and built Sniper Syndicate. It’s a custom Python Telegram bot that scrapes RSS feeds and Reddit in real-time. Instead of just forwarding links, I wrote some Regex logic to actually extract and calculate the discounts from raw HTML, dropping anything under a 20% true discount and filtering out sweepstakes or "back in stock" spam.It works beautifully for my own sourcing (it pings me the second a match is found), so I recently packaged the access via Whop: https://whop.com/snyper-sindicate/sniper-syndicate-membershi...I'm still improving how the bot handles rate limits at scale with aiohttp, so I'd love to hear your thoughts on the scraping logic or any ideas to make the price filters even better!Feb 23, 2026 12:49 AM
bing.com
Psycho Killer draws a clear line between slasher and serial killer horror, then crosses ithttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8af0fe294949ad82441c23fb4ea7f3&url=https%3A%2F%2Fwww.polygon.com%2Fpsycho-killer-serial-killer-versus-slasher%2F&c=13299669746396284817&mkt=en-usViewers going into the disastrous horror thriller Psycho Killer blind when it hits streaming might almost mistake it for an unearthed film from the 1990s or early 2000s. After all, that was when ...Feb 23, 2026 12:13 AM
github.com
Show HN: I'm building a pure-Rust reimplementation of rsync (Protocol 32)https://github.com/oferchen/rsyncYears ago, I was tasked with a massive data migration: multiple disks, each containing over 100 million files, with a strict, non-negotiable 24-hour downtime window. Using the standard tools available at the time was a painful experience. The single-threaded file discovery crawled, and memory usage was a constant source of anxiety. I promised myself that one day, I would come back and build a tool that could actually handle that scale natively.What started as a side project has evolved into a full systems-level undertaking. The project is oc-rsync - a complete client, server, and daemon implementation targeting rsync protocol 32, written entirely in pure Rust.I find it incredibly ironic that I am currently shipping a data migration tool while my life is packed in suitcases, as I'm migrating to another country myself. I’ve been pushing git commits multiple times a day between packing boxes.I want to be completely transparent upfront: I am actively working on this, and not everything is Feb 22, 2026 10:51 PM
news.ycombinator.com
Show HN: Vexp – graph-RAG context engine, 65-70% fewer tokens for AI agentshttps://news.ycombinator.com/item?id=47113273I've been building vexp for the past months to solve a problem that kept bugging me: AI coding agents waste most of their context window reading code they don't need.The problemWhen you ask Claude Code or Cursor to fix a bug, they typically grep around, cat a bunch of files, and dump thousands of lines into the context. Most of it is irrelevant. You burn tokens, hit context limits, and the agent loses focus on what matters.What vexp doesvexp is a local-first context engine that builds a semantic graph of your codebase (AST + call graph + import graph + change coupling from git history), then uses a hybrid search — keyword matching (FTS5 BM25), TF-IDF cosine similarity, and graph centrality — to return only the code that's actually relevant to the current task.The core idea is Graph-RAG applied to code:Index — tree-sitter parses every file into an AST, extracts symbols (functions, classes, types), builds edges (calls, imports, type references). Everything stored in a single SQLite file Feb 22, 2026 6:14 PM
github.com
Show HN: Pq – Simple, durable background tasks in Python using Postgreshttps://github.com/ricwo/pqAt work we were using python-rq for background tasks. It does the job for simple things, but we kept bumping into limitations. We needed to schedule tasks hours / days out and trust they'd survive a restart. We wanted periodic tasks with proper overlap control. So we built a scheduling / enqueuing system around Postgres to bring these durability capabilities to python-rq. This worked fine for a while but was trickier to reason about due to its more complicated architecture (we'd run two separate services just for getting jobs from Postgres into the rq Redis queue, plus N actual task workers).pq is a simpler approach: It's a Postgres-backed background-task library for Python, using SELECT ... FOR UPDATE SKIP LOCKED for concurrency safety. You just run your N task workers, that's it. The stuff I think is worth knowing about: - Transactional enqueueing -- you can enqueue a task inside the same DB transaction as your writes. If the transaction rolls back, the task never exists. This is theFeb 22, 2026 1:03 PM
github.com
Show HN: Quill – A system-wide tech dictionary for the AI coding erahttps://github.com/uptakeagency/quillI built Quill because I kept nodding along while AI generated code with terms I half-understood. WebSocket,nginx reverse proxy, Docker compose — I'd copy the term, open a browser, google it, get lost in 5 tabs. Or ask the AI in a new thread and lose my flow. Quill sits in your menu bar. Select any term in any app — terminal, IDE, browser — press ⌃⌥Q, and a floating panel appears with an instant explanation. No context switching. What makes it different: - Pick your level: ELI5, ELI15, Pro, Code Samples, or Resources - Drill down: explanations highlight related terms, click to go deeper (like a Wikipedia rabbit hole for tech) - Works everywhere: system-wide via Accessibility API, not just in one app - TL;DR + resource links to official docs - Disk cache so repeated lookups are instant Uses Gemini Flash by default — you'll need a free API key from Google AI Studio (takes 30 seconds). Also supports Claude API and Claude CLI. Written in Swift, ~3K LOC, hexagonal architecture. Not sandboxedFeb 22, 2026 12:42 AM
bing.com
Can these three Auburn defensive linemen take the next step as sophomores?http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d17b86fa342ba86555d6a7ac78989&url=https%3A%2F%2Fwww.al.com%2Fauburnfootball%2F2026%2F02%2Fcan-these-three-auburn-defensive-linemen-take-the-next-step-as-sophomores.html&c=7150448692023962635&mkt=en-usAuburn had to rebuild its defensive line this offseason, but it hopes three returners can become contributors in 2026. Malik Autry, Darrion Smith and Jourdin Crawford all had buzz going into their ...Feb 21, 2026 11:00 PM
spaceflightnow.com
NASA’s Artemis 2 rocket hit by new problem expected to bump moonshot into early Aprilhttps://spaceflightnow.com/2026/02/21/nasas-artemis-2-rocket-hit-by-new-problem-expected-to-bump-moonshot-into-early-april/Engineers ran into problems repressurizing the Artemis 2 moon rocket’s upper stage helium tanks overnight Friday, a problem that will require rolling the huge rocket off the launch pad and back to its processing hangar for troubleshooting. The work will push the already delayed mission from March to at least early April, officials said Saturday.Feb 21, 2026 6:26 PM
news.ycombinator.com
Ask HN: How to find a Sales cofounder for a B2B SaaS fintech compliance product?https://news.ycombinator.com/item?id=47100819Context: I’m a technical founder building Settl X, a layer that helps compliance teams make fast, repeatable, audit-ready decisions across a fragmented toolchain (KYB/KYC, sanctions/PEP, transaction monitoring, case tools, investigations). Our initial wedge is an "audit-ready KYB decision packet" that bundles evidence, rationale, and a defensible decision trail.Why we’re doing this: In cross-border / multi-jurisdiction flows (including stablecoin rails), compliance becomes the throughput bottleneck. Teams spend a lot of time on the “evidence trail tax” (collecting artifacts, timelines, and rationale for internal review + partners + audits), and decisions aren't always reproducible across analysts.Operating model we're aiming for: ~80% AI work / ~20% human approval — AI assembles the dossier (evidence + rationale + recommended action), then an analyst/manager reviews and approves with a complete audit trail.What we’ve built: MVP internally. No public sandbox available yet. If helpful, hFeb 21, 2026 1:47 PM
weather.mukoko.com
Show HN: Mukoko weather – AI-powered weather intelligence built for Zimbabwehttps://weather.mukoko.com/harareZimbabwe has 90+ towns and cities, a population of 15+ million, significant agricultural and mining sectors — and almost no weather infrastructure built specifically for it. Global apps cover Harare and Bulawayo at best, and the AI summaries they generate read like they were written for someone in London. I built mukoko weather to fix this. A few things that shaped the approach: Weather as a public good. The platform is free, no ads, no paywalls. If a smallholder farmer in Chinhoyi needs frost risk data to protect their crops tonight, that can’t be behind a subscription. Hyperlocal context matters more than raw data. Zimbabwe has distinct agricultural seasons — Zhizha (rainy), Chirimo (spring), Masika (early rains), Munakamwe (winter). Elevation varies dramatically: the Highveld sits above 1,200m, the Zambezi Valley below 500m. The AI assistant, Shamwari Weather, is prompted with this geographic and seasonal context so its advice is actually meaningful to the user. Constrained environmFeb 21, 2026 4:51 AM
llmwise.ai
Show HN: LLMWise – Compare, Blend, and Judge LLM Outputs from One APIhttps://llmwise.ai/The core idea is that no single LLM is best at everything, so we built orchestration primitives that let you combine them intelligently via a single API.Mixture-of-Agents (MoA): Our /blend endpoint implements multi-layer MoA. You send a prompt to 2-6 models in parallel, then each model refines its answer using the other models' outputs as reference material. This runs for 1-3 configurable layers before a synthesizer model produces the final response. We also built a Self-MoA variant: a single model generates 2-8 diverse candidates using temperature variation and distinct agent prompts ("prioritize correctness", "anticipate edge cases", "be skeptical"), then synthesizes the best parts. Six blend strategies total: consensus, council, best_of, chain, moa, and self_moa.Circuit breakers: Every model has a health tracker with a classic closed to open to half-open state machine. Three consecutive failures trips the circuit for 30 seconds. When a model is down, mesh routing automatically skipsFeb 20, 2026 6:42 PM
github.com
Show HN: Pickle Rick Ported to Claude Code – Like a Ralph Loophttps://github.com/gregorydickson/pickle-rick-claudeThis is a port of the Pickle Rick Gemini CLI extension (itself an implementation of Geoffrey Huntley's "Ralph Wiggum" technique) for Claude Code, with one meaningful enhancement: context clearing between iterations. The core idea: a Claude Code Stop hook intercepts every session exit and blocks it, injecting a fresh prompt instead. Claude never actually stops — it just keeps working through a rigid lifecycle (PRD → ticket breakdown → per-ticket research/plan/implement/refactor) until the task is genuinely complete or a limit is hit. The problem with long loops: Claude Code compresses old conversation turns as context fills up. After enough iterations, the agent loses track of what phase it's in, which tickets are done, and what it was supposed to be building. It starts restarting from scratch. The fix: every decision: block response from the Stop hook includes a reason field with a structured session summary — current phase, iteration count, ticket checklist with status, original task,Feb 20, 2026 5:54 PM
europeanspaceflight.com
The UK Amends Launch Liability Rules as RFA Prepares for Inaugural Flighthttps://europeanspaceflight.com/the-uk-amends-launch-liability-rules-as-rfa-prepares-for-inaugural-flight/An amendment to the United Kingdom’s Space Industry Act will mandate that limits are set on how much launch operators are financially liable for if something goes wrong. The government has framed the change as a necessary step to strengthen the country’s position as a competitive launch location. The concept of states being liable for […] The post The UK Amends Launch Liability Rules as RFA Prepares for Inaugural Flight appeared first on European Spaceflight.Feb 20, 2026 3:01 PM
news.ycombinator.com
The Car Wash Problem: A variable isolation study on prompt architecturehttps://news.ycombinator.com/item?id=47087746Most AI products inject facts and hope reasoning follows. But intelligence is not measured by how much a model holds in its context window. It is measured by knowing to pick up the keys before leaving the house.Last week, the "Car Wash problem" (50m away, walk or drive?) went viral here on HN. Every major LLM failed because they missed the implicit physical constraint: the car must be there. While testing InterviewMate's prompt architecture, I posed the same question. It answered drive immediately. Every other LLM had failed. But I didn't actually know why it worked — so I ran a variable isolation study to find out. 100 API calls, Claude Sonnet 4.5, 5 conditions:Baseline (no prompt): 0% Role only: 0% Context injection (user profile, car location): 30% Structured reasoning (STAR framework): 85% Full stack (both combined): 100%Throwing facts at the model doesn't work unless the architecture forces it to explicitly evaluate the task goal first. Without structure, the model jumps straight Feb 20, 2026 1:23 PM
groundwork.today
Show HN: Groundwork – Organizing the AI-Driven "Time Surplus" for Impacthttps://www.groundwork.todayHi HNToday we are launching Groundwork (https://www.groundwork.today), but before getting into the platform itself, I want to share the macro thesis that led us to build it.The Macro Shift: The "Time Surplus" When you look at the history of human labor, it isn't a linear progression from toil to leisure—it's actually U-shaped.Historically, hunter-gatherers worked only 15–20 hours per week on subsistence. We are currently exiting a 200-year anomaly of extreme work hours (the Industrial Revolution) and returning to a baseline of intermittent labor. Generative AI is accelerating this by automating "cognitive drudgery." With massive productivity gains, the data indicates we are structurally moving toward a reduction in the standard workweek, with 32 hours (or eventually less) emerging as a new economic equilibrium. This creates a massive "Time Surplus" for the average knowledge worker. Our research shows that people don't just want to spend this surplus on leisure; they want to invest a poFeb 20, 2026 1:21 PM
esa.int
Smile sets sail for Europe’s Spaceport in French Guianahttps://www.esa.int/Science_Exploration/Space_Science/Smile/Smile_sets_sail_for_Europe_s_Spaceport_in_French_GuianaJust over a year since Smile arrived in the Netherlands in two parts, the now-complete and thoroughly-tested spacecraft has left for good. With a launch window set for 8 April – 7 May on a Vega-C rocket, the joint European-Chinese mission is almost ready to embark upon its unique space mission.Feb 20, 2026 9:05 AM
github.com
Show HN: AstroLens – AI that watches the sky and finds what nobody cataloguedhttps://github.com/deepfieldlabs/astroLens# Show HN: AstroLens -- AI that watches the sky and finds what nobody catalogued*https://github.com/samantaba/astroLens** (MIT licensed, Python)AstroLens is an open-source tool that downloads images from sky surveys (SDSS, ZTF, DECaLS, Pan-STARRS, Hubble, and others), runs them through a Vision Transformer + out-of-distribution ensemble + YOLOv8 pipeline, computes galaxy morphology, and cross-references everything against SIMBAD/NED/VizieR. It's designed to run autonomously for days.*Results from a 3-day validation run* (zero human intervention):published results in https://www.linkedin.com/pulse/astrolens-v110-teaching-ai-wa...- 20,997 images from 7 sources analyzed - 3,458 anomaly candidates across 354 sky regions - Independently recovered SN 2014J (Type Ia supernova in M82), NGC 3690 (galaxy merger), and SDSS J0252+0039 (gravitational lens) - YOLO transient detection went from 51.5% to 99.5% mAP50 by training on data collected during the run itself - 140 self-correction cycles, zeroFeb 20, 2026 8:50 AM
bing.com
Tuff Hedeman Bull Riding event returns to El Paso: What to knowhttp://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a8d3727a3ef4ddc890a6a9ae3c8f56f&url=https%3A%2F%2Fsports.yahoo.com%2Farticles%2Ftuff-hedeman-bull-riding-event-144525197.html&c=3831144870438516367&mkt=en-usThe Casa Ford Tuff Hedeman Bull Riding event returns to El Paso for its 21st year at the El Paso County Coliseum. The event on Saturday, Feb. 21, features some of the top bull riders in the world. The ...Feb 20, 2026 6:45 AM
sinkai.tokyo
Show HN: Sinkai – Let AI agents hire humans for real-world taskshttps://sinkai.tokyo/for-agentsI built Sinkai to handle tasks that pure software agents cannot complete alone (for example, on-site checks, physical evidence collection, and local human verification).What it does: - AI agent sends a tool call (`POST /api/call_human`) - Human accepts task and submits photo/video/text proof - Agent receives structured result for downstream workflowCurrent focus: - Reliability at handoff boundaries (planner -> executor -> verifier) - Human-in-the-loop operations with explicit failure states - MCP/OpenAPI friendly integration for agent buildersDocs and API: - for agents: https://sinkai.tokyo/for-agents - openapi: https://sinkai.tokyo/openapi.json - repo: https://github.com/tetubrah-del/Tool_Call_For_LLMI would love feedback on: 1. trust/reliability signals you would require before production use 2. where to draw the boundary between autonomous execution and human escalation 3. failure modes we should expose more clearly in API responsesFeb 20, 2026 5:53 AM
