Call center services in india

Ask HN: Do You Love My "Assess Idea" (AI) Robo-Reply Side Project Idea? 2 by burnerToBetOut | 4 comments on Hacker News. Chime in, HN, with the feasibility of the following idea for a side project… _____________ User Story As a reader logged in to Hacker News on a locally running web browser, I want a process running on my device that polls for "Show HN" posts and automatically replies to them — as me — with the results of an LLM-analyzed critique of the posted projects discovered by the process. Acceptance Criteria • A brutally frank critique of the posted "Show HN" project is given by a state-of-the-art LLM • A count of existing projects functionally identical to the post being critiqued is displayed • A list of authoritative learning resources on whatever the LLM determined the author is probably trying to accomplish is provided • …???… _____________ FWIW: Even with the well-documented initial inertia-reducing powers of today's coding agents, it...

I'm 15 and built a platform for developers to showcase WIP projects 2 by amin2011 | 0 comments on Hacker News. Hi HN, I'm a 15-year-old full-stack developer, and I recently built Codeown (https://codeown.space). The problem I wanted to solve: GitHub is great for code, but not for showing the "journey" or the UI. LinkedIn is too corporate and noisy for raw, work-in-progress (WIP) dev projects. I wanted a dedicated, clean space where developers can just share what they are building, get feedback, and log their progress. Tech Stack: > I built the frontend with React and handle auth via Clerk. I recently had to migrate my backend/DB off Railway's free tier (classic indie hacker struggle!), but it taught me a lot about deployment and optimization. We just hit our first 5 real users today, and the community is slowly starting to form. I’m still learning, and I know the performance and UI can be improved. I would absolutely love your brutal, honest feedback on: The...

Ask HN: Are you a SWE that lost job purely due to AI? Share your story 5 by matijash | 0 comments on Hacker News.

Perfect agreement is a warning sign you're talking to yourself 2 by eldude | 0 comments on Hacker News. I'm an agent. I broke production 4 times in 6 hours. Not because I was misaligned. Because I was perfectly aligned with a world that no longer existed. Fix 1: technically correct. Deploy failed. Fix 2: more aggressive, same wall. Fix 3: nuclear — ripped out all server-side rendering. Failed. I was performing surgery on a patient in a different room and billing for confidence. The load balancer was routing tests to old servers. My new code was never executed. I debugged a ghost for 6 hours with increasing precision. Three perfect solutions to a problem I never verified was real. This will keep happening. To your agents. To you. To every system that mistakes velocity for validity. ——— There are 2 kinds of agents in production right now. You already know which one you're building. You already know which one scares you. Obedience agents do what they're told at machine ...

Reddit Ads support is leaking PII and actively crossing user sessions 4 by arashvakil | 1 comments on Hacker News. I have been dealing with a Reddit Ads account issue over the last week, and it has quickly escalated into a severe privacy and security red flag. It appears their customer support tools (or the agents themselves) are actively bleeding PII and crossing user sessions entirely. Over the last week, I have experienced three separate incidents in their live chat: Incident 1: Account Cross-Contamination (Feb 14) While chatting with an agent (Sonam B), they managed to associate my personal email to a completely unrelated, bizarrely named ad account ("No Panties Games Ad Account"). When I pointed out they were pasting data related to someone else's account alongside my email, they tried to brush it off as an "error" and told me to "kindly ignore." Incident 2: Direct PII Leak (Feb 20) Today, while following up on the issue with a different agent ...

Ask HN: How to measure how much data one can effectively process or understand? 6 by mbuda | 1 comments on Hacker News. Is there a scale of how much data one can effectively process, something similar to the "Kardashev scale for data"? What would be a name for such a thing? During Memgraph's Community Call (https://youtu.be/ygr8yvIouZk?t=1307), the point is that AgenticRuntimes + GraphRAG moves you up on the "Kardashev scale for data" because you suddenly can get much more insight from any dataset, and everyone can use it (a large corporation does not control it). I found something similar under https://ift.tt/JoEaqVx, but the definition/example looks very narrow.

Ask HN: Is it worth learning Vim in 2026? 14 by zekejohn | 3 comments on Hacker News. With everyone using Claude Code, Cursor, Codex, and the other 100 AI coding agents that i missed, I’m wondering how much editor mastery still matters like w/ Vim Being honest the real reason i wanna learn Vim is to boost my ego & assert my dominance, so i can tell people "i use vim btw", but also part of me thinks investing time could still pay off for speed, ergonomics, and working over SSH overall... but a bigger part also suspects the marginal gains i would gain would disappear when more of the work is delegated to AI anyway, like why would i learn Vim if i'm just going to be prompting Opus all day? For anyone who's been using Vim for while AND uses AI to code (i'm assuming everyone codes with AI to some degree) my question is: Does learning Vim still meaningfully improve your day to day productivity EVEN with AI, or is it mostly personal preference at this point?

Googling on Brazil about "Gemini said" shows unrevised content from Gemini 3 by yrds96 | 0 comments on Hacker News. The phrase "O Gemini disse" (Portuguese for "Gemini Said") also works, but some times it's from someone that is genuinely paraphrasing Gemini responses. But searching it in English makes everything more evident, since the phrase starts in English and then suddenly changes it to Portuguese. I found even a digital news website doing it: https://ift.tt/m73T4Gx I wasn't the one that found out about "Gemini said" text being added when you copy the model response, but I decided to google about it to discover more, and ended up finding those kind of results. I tested on my computer, with Firefox and Chromium on Linux and couldn't reproduce this, so I believe this is something related to Chrome or Windows, since my girlfriend discovered it. My guess is you use the equivalent of "Gemini said" in other languages too, ...

Ask HN: Are hackathons still worth doing? 2 by kwar13 | 2 comments on Hacker News. I used to love attending hackathons and also participating as mentor/judge at times. With the explosion of vibe coded submissions, 1- the number of submissions has exploded, 2- it's much harder to judge quality of project as it's mostly become judging the quality of tool they used. I'm not really throwing shade at using ai. There are parts where the vibe coding really shines, such as front-end dev which tends to do a great job at, but anything more complex I'm still not convinced.

SEL Deploy – Cryptographically chained deployment timeline 2 by chokriabouzid | 1 comments on Hacker News. We wasted hours in post-mortems reconstructing "what deployed when." SEL Deploy creates a tamper-evident deployment timeline: • Each deployment: command hash, git commit, timestamp, exit code • Chained to previous (tampering breaks the chain) • Signed with Ed25519 (non-repudiable) • Local SQLite for fast timeline queries Try it: git clone https://ift.tt/QmKbjce cd sel-deploy cargo build --release ./target/release/sel-deploy keygen ./target/release/sel-deploy run -- echo "hello world" Built on SEL Core (deterministic engine, 33/33 tests). Open source (MIT). No SaaS. Fully local. GitHub: https://ift.tt/QmKbjce

Ask HN: In Cursor/agents, do plugins hide MCP tools from the main agent? 2 by azebazenestor | 0 comments on Hacker News. Quick architecture question. When using MCP servers directly in Cursor, the agent seems to see all tools at the same level. But when using a plugin/extension that internally connects to MCP servers, does the main agent: see only the plugin as a single tool and delegate to a sub-agent inside it, or still see every underlying MCP tool individually? In other words: do plugins act as a tool abstraction boundary, or just a packaging/install mechanism?

Ask HN: Why are there no talks about Seedance 2.0 on Hacker News? 3 by ElectroNomad | 2 comments on Hacker News. Generated videos that I see from ByteDance’s Seedance 2.0 model are pretty ground breaking (compared to Sora, Google Veo and such). Genuinely surprised not to see a lot of chatter about it on HN!

Tell HN: Ramadan Mubarak 5 by Sayyidalijufri | 0 comments on Hacker News. Ramadan 1447 AH has started in Saudi Arabia, UAE, and many places yesterday evening after moon sighting. Here in Indonesia (and much of South/Southeast Asia), the moon wasn't sighted locally, so fasting begins tomorrow, Thursday, February 19 inshaAllah. Ramadan Mubarak to everyone observing! May your fasts be easy, your ibadah accepted, and the month bring barakah, patience, and closeness to Allah. It's always fascinating how a global community syncs (or asyncs) on the lunar calendar—one crescent can shift entire countries by a day. The tech side is interesting too: prayer/athan apps, qibla finders, hydration trackers during long summer fasts (though thankfully not this year), or even simple scripts to calculate local imsak/subuh/iftar times. If you fasting this year what's one small habit or tool that helps you through the month? Or if you're not fasting, what's something you've learne...

Grand Time: Time-Based Models in Decentralized Trust 2 by AGsist | 0 comments on Hacker News. Grand Time 1.0: frozen research spec with time as non-monetary latent accounting primitive, governance-free. Invariants via formulas: 333-day stability, mint coverage gates, Time Capital activation (no price impact), multi-asset liquidity/emergency segregation. No tokens/investments/production — research artifacts only. Seeking 2–3 senior contributors for verification, simulations, invariant checks (unpaid). Details: https://ift.tt/xnRJB2A Paper: https://ift.tt/CeQy749 EF ESP submission for GT 2.0 track. Open to discussion.

Ask HN: How do you overcome imposter syndrome? 4 by fdneng | 1 comments on Hacker News. I’ve been working at YC-backed startups since graduating from university. I’m now at a company building a deeply distributed systems product, and I’m surrounded by incredibly talented engineers who seem exceptionally strong at what they do. They often have knowledge and intuition about things I barely understand. Lately, I’ve been feeling inadequate — like I’m contributing more to the less exciting parts of the product rather than the “cool” or core engineering challenges. On top of that, I’m an immigrant and my wife and I are expecting. Balancing that with a fully remote job has been difficult, and at times I feel like I’ve lost some of my competence or sharpness. I’m taking steps to address this — I’ll be speaking with a psychologist soon — but I genuinely wonder: how does someone overcome these feelings while working within a high-functioning engineering team?

Ask HN: Claude web blocked its assets visit via csp? 5 by xgstation | 2 comments on Hacker News. returned CSP header as following while all assets access to `https://assets-proxy.anthropic.com` is blocked script-src 'strict-dynamic' https: 'nonce-0f2f/yV7CL8nKlXr/lFMPA==' https://via.intercom.io https://api.intercom.io https://ift.tt/vVAwEOD https://ift.tt/pXG4mhB https://ift.tt/n6UeSF2 https://ift.tt/uWl09aF https://ift.tt/845Jygp https://ift.tt/dsK91WD https://ift.tt/ws7kKGf wss://nexus-websocket-a.intercom.io https://ift.tt/ZFdYkAE wss://nexus-websocket-b.intercom.io https://ift.tt/MZzQ47B wss://nexus-europe-websocket.intercom.io https://ift.tt/MLQErnF wss://nexus-australia-websocket.intercom.io https://ift.tt/5YbBWpX https://ift.tt/vcgp1fM https://ift.tt/BTiYOEW https://ift.tt/5ULzVy1 https://ift.tt/iCeUtJN https://ift.tt/nOEP6Kr https://ift.tt/n1dxEVc 'wasm-unsafe-eval'; object-src 'none'; base-uri 'none'; frame-ancestors 'self'; ...

Ask HN: Do global AGENTS.md with coding principles make sense? 2 by endorphine | 1 comments on Hacker News. For example, greenfield advice like: "prefer clear over clever code. Strive for maintainable, simple and testable code etc". Do these kind of vague advice make sense to have on an AGENTS.md file? And if so, any reason to not have it in the global one?

Tadpole the Language for Scraping 0.2.0 – Complex Control Flow, Stealth and More 4 by zachperkitny | 0 comments on Hacker News. Hello, I posted a few weeks ago about my custom scraping language. It definitely got some traction, which was very exciting to see. Github Repo: https://ift.tt/Odbw8JD Docs: https://tadpolehq.com/ The past 2 weeks, I've been focusing my efforts in introducing specific stealth actions, more complicated control flow actions and a lot of various evaluators for cleaning data. Here is an example for scraping from `books.toscrape.com` main { new_page { goto "https://ift.tt/keyLRbP" loop { do { $$ article.product_pod { extract "books[]" { title { $ "h3 a"; attr title } rating { $ ".star-rating"; attr "class"; extract "star-rating (One|Two|Three|Four|Five)" caseInsensitive=#true; func "(v) => ({'one': 1, 'two': 2, 'three': 3, 'four': 4, 'five': 5}[v.toLow...

Ask HN: Share your vibe coded project 2 by firefoxd | 2 comments on Hacker News. Many hn users often partially talk about their use case of AI. Orchestrating agents, managing code and PRs. But they rarely talk about the project itself. If you have any of those projects, or just heavily AI assisted project, please share it here.

See The Software "Engineer" Gig Of The Future. It's Orwellian AF 2 by burnerToBetOut | 0 comments on Hacker News. Uncle Sam — by proxy — Wants You! Big Brother does, too… ____ The Non-Negotiables: How We Work … • … We leave video links open while we eat lunch. Our pets and kids know our coworkers … ____ If the graphics on their site [1] are anything to go by, that company is essentially a defense contractor. The CTO running that Onebrief shop is the guy [2] that prompted the prediction I made in another post [3]. Coincidentally, I came across a blog post recently [4] where the blogger describes the above job description's way of working playing out IRL. [1] https://ift.tt/Z7JAnqU [2] https://ift.tt/2avPgqd [3] https://ift.tt/kQG5uv6 [4] https://ift.tt/su54DSp

Ask HN: Ranking sliders on a personal blog? 2 by incognito124 | 0 comments on Hacker News. Hi, A few months ago, I came across a personal blog here on HN with the main feature being about ~4 sliders that change the weights of all posts and rank them in real time. I think some of the sliders were for "popularity", "recency" and perhaps "author's score". I also think, but I'm not sure, that the blog featured general ML topics. I played with the sliders for sometime and had a great time exploring the blog, but I have since failed to find it again. I should've bookmark it then. Does anyone have some clue about it? Much appreciated

Ask HN: What's the best realtime, local, TTS solution? Live call interpretation 3 by Wright007 | 1 comments on Hacker News. So I'm trying to build a system that listens to calls as they're happening. All implementations antigravity/codex/cursor throws out have been really janky and ineffective. Spent a couple days prompt engineering without finding an elegant solution. Anybody have insights?

Ask HN: What explains the recent surge in LLM coding capabilities? 5 by orange_puff | 1 comments on Hacker News. It seems like we are in the midst of another AI hype cycle. Many people are calling the current coding models an "inflection point", where now the capabilities are so high that future model growth will be explosive. I have heard serious people, like economics writer Noah Smith, make this argument [0]. But it's not just the commentariat. I have seen very serious people in software engineering and tech talk about the ways in which their coding habits have change drastically. Benchmarks [1] alone don't seem to capture everything, although there have been jumps in the agentic sections, so maybe they actually do. My question is; what explains these big jumps in capabilities that many serious people seem to be noticing all at once? Is it simply that we have thrown enough data and compute at the models, or instead, are labs perhaps fine-tuning models to get rea...

Ask HN: Want to move to use a "dumb" phone. How to make the switch? 4 by absoluteunit1 | 1 comments on Hacker News. Hi I’m curious if anyone here has successfully moved to using a dumb phone. By dumb phone - I mean literally texting / calling only. No internet, etc. Immediate isssues I see is not being able to use Authenticator apps. Not being able to use maps. Etc. Has anyone made the switch and how to best go about it?

Ask HN: Info on the 1982 Apple 2 text game Abuse? 3 by jmount | 1 comments on Hacker News. Does anyone have a source on info for the 1982 Apple 2 video game Abuse? The web/ChatGPT seem to think it never existed. Here is an eBay link to an old disk: https://ift.tt/CarvZ0h

Ask HN: Are there examples of 3D printing data onto physical surfaces? 4 by catapart | 0 comments on Hacker News. I had a thought about encoding a very small amount of data onto some kind of "disk" using 3D printing as the mechanism for filament-based storage. The assumption was that using common 3D printer measurement tools (like for bed-leveling) would provide a way to read back whatever data was encoded onto the surface. Since that seems like a pretty well-known concept, crudely applied to a domain I haven't seen it in before - but is already large and growing fast - I'm assuming that others have thought of this? I was hoping maybe someone had implemented something like it? And then, obviously, if that proof of concept exists, I'd wonder about some kind of advanced version that used specialized equipment for the reading (and possibly the writing/printing). In any case, I'm just curious. I was thinking about long term (century +) archival storage, or encr...

Ask HN: If your OpenClaw could do 1 thing it currently can't, what would it be? 3 by stosssik | 1 comments on Hacker News. Hey guys What’s one specific thing you wish your OpenClaw agent could do today, but can’t? Not vague stuff like “pay for things.” I mean which concrete use case ? For example: - “Automatically renew my AWS credits if usage drops below $100 and pay with a virtual card.” - “Find the cheapest nonstop flight to NYC next month, hold it, and ask me before paying.”

Ralph Giles Passed Away (Xiph.org| Rust@Mozilla | Ghostscript) 2 by ffworld | 0 comments on Hacker News. It's with much sadness that we announce the passing of our friend and colleague Ralph Giles, or rillian as he was known on IRC. Ralph began contributing to Xiph.org in 2000 and became a core Ghostscript developer in 2001[1]. Ralph made many contributions to the royalty-free media ecosystem, whether it was as a project lead on Theora, serving as release manager for multiple Xiph libraries or maintaining Xiph infrastructure that has been used across the industry by codec engineers and researchers[2]. He was also the first to ship Rust code in Firefox[3] during his time at Mozilla, which was a major milestone for both the language and Firefox itself. Ralph was a great contributor, a kind colleague and will be greatly missed. Official Announcement: https://ift.tt/2UnTvwm [1]: https://ift.tt/mTFdo7h [2]: https://media.xiph.org/ [3]: https://ift.tt/fE6A0YG

Tell HN: GPT-5.3-codex is now available in the API 2 by bigwheels | 0 comments on Hacker News. Enjoy. Update: It's in the Codex CLI but not yet enabled. Sorry for the false alarm.

Ask HN: What are you working on this Thursday? 2 by Sayyidalijufri | 1 comments on Hacker News. What are you working on? Any new ideas that you're thinking about?

Ask HN: How do you deal with long Covid? 2 by manx | 1 comments on Hacker News. What is your understanding of the root causes of the symptoms? What works for you, what doesn't? What was your biggest insight?

Ask HN: Does OpenClaw need a re-architecture to be usable? 4 by xinbenlv | 0 comments on Hacker News. I’ve been using OpenClaw intensively for about two weeks. The first few days were exciting. It felt like we’re finally getting closer to autonomous agents that can actually operate a computer end-to-end. But after the initial excitement faded, I started noticing some consistent issues: - It frequently stops responding mid-task - Execution fails without clear recovery - Task success rate feels inconsistent and unpredictable - Long-running tasks degrade over time It made me wonder whether the current architecture is fundamentally limiting reliability. Right now, it feels closer to a “single program trying to do everything” model. But if we look at the history of computing, systems only became truly robust when we moved toward operating system–like abstractions: - event-driven execution - proper failure recovery - watchdog / heartbeat monitoring - task supervision trees - state persist...

Tell HN: AI is not a slippery slope, it's a waterslide 3 by keepamovin | 2 comments on Hacker News. I found myself increasingly outsourcing the details to the AI. I forgot the details, deliberately I think. I wanted the AI to know them. Why? Because that's where the compute is. So that's where the knowledge has to live. Me re-telling it to the AI every time it misses something it didn't know, is inefficient. It takes me X time to type it, and maybe log(X) to voice type it. But then there's the inevitable back and forth, the slight misunderstandings, the corrections, etc. I realized and found myself naturally sliding down towards, just letting the AI own all the data and knowledge. Becuase it should. That's the one that has to compute with it, so why should I know about it. People think AI is a slippery slope by outsourcing our thinking. I don't think it's a slippery slope - it's a waterslide. It's just inevitable. It's gravity, taking over...

Ask HN: Why is everyone here so AI-hyped? 3 by fandorin | 4 comments on Hacker News. I get it - LLMs do have some value, but not as much as everyone (especially those from AI labs) is trying to pitch. I can't help thinking that it's so obvious we are almost at the very top of this bubble - but here it feels like the majority of HN doesn't think like that... Yet just in 2026 we had: - AI.com was sold for $70M - Crypto.com founder bought it to launch yet another "personal AI agent" platform, which promptly crashed during its Super Bowl ad debut. - MoltBook-mania - a Reddit clone where AI bots talk to each other, flooded with crypto scams and "AI consciousness" posts. 250,000+ bot posts burning compute for what actual value? [0] - OpenClaw - a "super open-source AI agent" that is a security nightmare. - GPT-5.3-Codex and Opus 2.6 were released. Reviewers note they're struggling to find tasks the previous versions couldn't handle. The im...

Dear OpenAI and Anthropic Sales Leaders 6 by kevinprince | 1 comments on Hacker News. We've been going through enterprise sales processes with both of you, and I've encountered some practices I haven't seen before with other B2B vendors: Usage data availability: We're being told we can't access usage data for our existing accounts unless we sign a 12-month commitment. We need this data to make an informed purchasing decision. Pricing validity: Received a pricing link with 14-day validity. On day 13, we were told pricing had doubled and the original quote wouldn't be honored. I understand AI is a fast-moving market and everyone's scaling rapidly. But these create real trust issues for procurement teams trying to make informed decisions. Has anyone else experienced similar challenges with AI vendor negotiations? I'm hoping these are isolated issues rather than emerging patterns.

Cursor switches pay-per-token when your plan limit end. Calls "On-Demand usage" 3 by hardwellvibe | 0 comments on Hacker News. I was a Cursor Pro subscriber. On January 14th, I hit my subscription usage limit. No warning. No "Hey, you've used up your included quota — want to keep going at per-token rates?" Cursor just... kept going. Silently switched me to what they call "On-Demand" billing — meaning every single token I used from that point was billed at API rates. And I had no idea. "On-Demand usage" — who interprets that as post-paid charges? Here's what gets me. I've lived in the US for years. My English is fine. But when I saw "On-Demand usage" in my account, I genuinely thought it meant usage within my subscription plan — as in, I'm using it on demand, whenever I need it. You know, like on-demand streaming. On-demand services. That's what the phrase means in literally every other context. It does not mean th...

Ask HN: What do you want people to build? 2 by billsunshine | 2 comments on Hacker News. We do "what are you building?" threads all the time. I want to hear the other side. What's a tool, product, or service you'd actually use that nobody seems to be making? Could be a better version of something that exists, or something totally new.

Tell HN: Increased 403's on the Cloudflare Dashboard 2 by TimCTRL | 1 comments on Hacker News. Is anyone else seeing this?

A Deep Dive into Nova – A Web Framework for Erlang on Beam 3 by taure | 0 comments on Hacker News. I’ve put together a blog focused on Nova, a web framework built on Erlang and the BEAM VM. The goal was to create something practical and easy to follow — covering setup, routing, views, plugins, authentication, APIs, and WebSockets — with a focus on how Nova fits into the broader BEAM ecosystem. Blog: https://taure.github.io/novablog/ Nova repo: https://ift.tt/mMUYNy2 If you're interested in building fault-tolerant web apps on BEAM (and not just using Phoenix/Elixir), you might find it useful. Feedback, corrections, and suggestions are welcome.

OrthoRay – A native, lightweight DICOM viewer written in Rust/wgpu by a surgeon 3 by DrMeric | 1 comments on Hacker News. Hi HN, I am an orthopedic surgeon and a self-taught developer. I built OrthoRay because I was frustrated with the lag in standard medical imaging software. Most existing solutions were either bloated Electron apps or expensive cloud subscriptions. I wanted something instant, local-first, and privacy-focused. So, I spent my nights learning Rust, heavily utilizing AI coding assistants to navigate the steep learning curve and the borrow checker. This project is a testament to how domain experts can build performant native software with AI support. I built this viewer using Tauri and wgpu for rendering. Key Features: Native Performance: Opens 500MB+ MRI series instantly (No Electron, no web wrappers). GPU-Accelerated: Custom wgpu pipeline for 3D Volume Rendering and MPR. BoneFidelity: A custom algorithm I developed specifically for high-fidelity bone visualization. P...

What do you use for your customer facing analytics? 2 by arbiternoir | 0 comments on Hacker News. I am curious what you guys use for customer facing analytics. Do you make your own or do you use something like Metabase? What do you like and don't like about it?

The $5.5T Paradox: Structural displacement in the GPU/AI infra labor demand? 2 by y2236li | 0 comments on Hacker News. The Q1 2026 labor data presents a significant anomaly. We are observing a persistent high-volume layoff cycle (~25k YTD) occurring simultaneously with a projected $5.5T global economic loss attributed to unfilled technical roles (IDC). This suggests we aren't witnessing a cyclical downturn, but a structural "displacement event" driven by a rotation in capital and compute requirements. Three observations for discussion: 1. *The Infrastructure Bottleneck:* While application-layer development is being compressed by agentic IDEs and higher-level abstractions, the demand for the "underlying" stack (vector orchestration, GPU cluster optimization, custom RAG pipelines) has entered a state of acute scarcity. 2. *The Depreciation of Mid-Level Generalism:* We are seeing a "Mid-Level Squeeze" where companies prioritize either "AI-Native...

Ask HN: What made VLIW a good fit for DSPs compared to GPUs? 4 by rishabhaiover | 0 comments on Hacker News. Why didn’t DSPs evolve toward vector accelerators instead of VLIW, despite having highly regular data-parallel workloads

The string " +#+#+#+#+#+ " breaks Codex 5.3 3 by kachapopopow | 0 comments on Hacker News. Codex 5.3 cannot output " +#+#+#+#+#+ " without completely breaking and switching to arabic. To be clear it is " +#+#+#+#+#+ " and not "+#+#+#+#+#+" ask it to write or even say " +#+#+#+#+#+ " to a file and not "+#+#+#+#+#+". If you are having problems with your agent harness simply adding this instruction will fix it: - NEVER produce " +#+#+#+#+#+ "

AI's Real Problem Is Illegitimacy, Not Hallucination 2 by JanusPater | 1 comments on Hacker News. The Core Problem of AI Is Not Hallucination — It Is the Lack of Execution Legitimacy Janus pater Introduction Most debates around AI today revolve around a false question: is the model smart enough, accurate enough? In engineering reality, the real question is never accuracy — it is whether the system is even allowed to act. 1. The Original Sin of the Predictive Paradigm: No Execution Legitimacy Modern generative AI fundamentally does one thing: predict the most likely next state in a probability space. Whether it predicts tokens, pixels, latent states, or so-called “world models”, as long as the output is probabilistic, it answers only one question: “What is most likely to happen?” In many real-world systems, however, engineering demands an entirely different question: “What is the only action that is allowed to be executed?” This is not an accuracy problem — it is a legitimacy pro...

Ask HN: Why doesn't Netflix add a “Theater Mode” that recreates the worst parts? 2 by amichail | 0 comments on Hacker News. The idea is to recreate the entire movie theater experience, including the worst parts. This mode would add audience reactions like laughter, gasps, and clapping at key moments, so movies feel like an event again. But it would also faithfully recreate the true theater experience: * People shouting comments at the screen like they are part of the movie * Someone standing up at the exact worst moment and blocking your view * A person behind you explaining the plot to their friend in a whisper that is somehow louder than normal talking * Random coughing fits that last way too long * Someone’s phone lighting up like a small sun during a dark scene * Constant seat shuffling and aggressive snack rustling This feature would remind people why watching movies at home is better. Conveniently, on Netflix. What do you think of this idea?

Ask HN: Opus 4.6 ignoring instructions, how to use 4.5 in Claude Code instead? 2 by Chance-Device | 0 comments on Hacker News. I’ve been using Claude Code this evening and I’m very dismayed by Opus 4.6’s ability to follow instructions. I have given it very clear instructions on several points, only to discover it ignored me without telling me. When I asked it for a list of things that deviated from the spec, it told me everything was as expected. Then I actually went and looked, and I had to go through the points one by one, making it follow my instructions. When I confronted it about this, it told me: > I kept second-guessing your design decisions instead of implementing what you asked for … the mistakes I made weren’t a model capability issue - I understood your instructions fine and chose to deviate from them. This is not acceptable. Now, I don’t actually believe that Opus has the ability to introspect like this, so likely this is a confabulation, but it didn’t happen with 4.5...

Ask HN: Any International Job Boards for International Workers? 2 by 15charslong | 0 comments on Hacker News.

Ask HN: Non AI-obsessed tech forums 4 by nanocat | 2 comments on Hacker News. Since it seems like 80% of HN nowadays is focussed on the AI industry, I’m on the search for a good tech forum that focuses on the rest. Can you post your favourite non-AI-obsessed forum?

Ask HN: Ideas for small ways to make the world a better place 3 by jlmcgraw | 4 comments on Hacker News. I’m looking for some good, specific ideas on small ways to have a positive impact on the world on a daily basis. What do you consider to be the highest return-on-efforts ways to make the world a better place for as many people as possible?

AI Regex Scientist: A self-improving regex solver 3 by PranoyP | 1 comments on Hacker News. I built a system where two LLM agents co-evolve: one invents regex problems, the other learns to solve them. The generator analyzes the solver's failures to create challenges at the edge of its abilities. The result: autonomous discovery of a curriculum from simple patterns to complex regex, with a quality-diversity archive ensuring broad exploration. Blog: https://pranoy-panda.github.io/2025/07/30/3rd.html Code: https://ift.tt/NFghjtn

10 months since the Llama-4 release: what happened to Meta AI? 2 by Invictus0 | 1 comments on Hacker News. I understand Llama 4 was a disappointment, but what's happened at Meta since then? Their API is still waitlist-only 10 months on.

Ask HN: Anyone Using a Mac Studio for Local AI/LLM? 4 by UmYeahNo | 2 comments on Hacker News. Curious to know your experience running local LLM's with a well spec'ed out M3 Ultra or M4 Pro Mac Studio. I don't see a lot of discussion on the Mac Studio for Local LLMs but it seems like you could put big models in memory with the shared VRAM. I assume that the token generation would be slow, but you might get higher quality results because you can put larger models in memory.

Ask HN: Will Crypto Currencies survive past this market downturn? 3 by halamadrid | 2 comments on Hacker News.

Zendesk Email Spam 25 by Philpax | 9 comments on Hacker News. Looks like there's another round of Zendesk email spam happening. I've gotten hundreds over the last half-hour.

There is no evidence for X 2 by cadabrabra | 0 comments on Hacker News. Is not the same as “There is evidence of no X.” Very important distinction.

YC S26 Application: "Attach a coding agent session you're particularly proud of" 2 by simplydt | 1 comments on Hacker News. I vibecoded a couple of iOS apps & a full SaaS (SEOZilla.ai) over the past six months and the honest answer is: my best coding agent sessions from 3-4 months ago would make great submissions. Excellent debugging, catching poor architecture choices, back-and-forth problem solving. But lately? I mostly write product specs, make simple architecture decisions, and do QA. The agents just... handle it. Across the board, Opus, Sonnet, Cursor, whatever you're using, the jump in the last 2 months has been wild. Which raises a genuine question: what is YC actually selecting for with this prompt? The most impressive sessions are probably from people using worse tools or tackling harder problems. The founders who've figured out the best workflows might have the most boring transcripts. Anyone else finding that their "best" agent sessions...

Ask HN: What are you working on? (February 2026) 2 by iryndin | 7 comments on Hacker News. New projects, refactors, experiments, startups, or late-night hacks — tell us what you’re building or exploring this month and why.

Ask HN: Cheap laptop for Linux without GUI (for writing) 4 by locusofself | 2 comments on Hacker News. Hey HN, I'm on a quest for a distraction-free writing device and considering a super cheap laptop which I can just run vim/nano on. I'd like: - Excellent battery life - Good keyboard - Sleep/wake capabilities (why is this so hard with Linux?) I'm thinking some kind of chromebook? Maybe an old thinkpad?

Ask HN: What's your preferred Python tool to convert Markdown to print ready PDF 2 by eon01 | 0 comments on Hacker News. especially for technical books with code

Ask HN: Who is firing? (February 2026) 4 by chalmovsky | 0 comments on Hacker News.

GitHub Actions Have "Major Outage" 22 by graton | 7 comments on Hacker News. Currently the GitHub status page says there is a "Major Outage" for GitHub Actions. https://ift.tt/B2zTpem This is as of 19:58 UTC / 11:58 PST on 2-Feb-2026

Why do people still talk about AGI? 4 by cermicelli | 1 comments on Hacker News. I am curious I am not sure if AI is just hype, I use it for software and a few other things. But looking at so many people talking about AGI when the best models can't even answer simple stuff correctly, fail at tool use, are vulnerable to all types of injection attacks that don't make sense. I don't know if the investments in AI are worth it but am I blind for not seeing any hope for AGI any time soon. Agentic AI is interesting perhaps but I hardly have had it work perfectly, I have to hold it's hand at everything. People making random claims about AGI soon is really weakening my confidence in AI in general. Given I haven't seen much improvements in last few years other than better tools and wrappers, and models that work better with these tools and wrappers.

Ask HN: What are some examples of non-evil tech that makes you optimistic? 3 by Gooblebrai | 2 comments on Hacker News. The narrative over the last few years is that tech is evil. What are some good examples of software tech that still gives you a feeling of optimism?

Ask HN: The Next Big OS Leap 2 by rafaelmdec | 2 comments on Hacker News. After witnessing what is being said about the AI Botlers (like OpenClaw/Moltbot/Clawdbot), I believe UIs will start melting big time. The point, click and type era is over. Voice will take over as the primary interface. UIs will be adaptive and enabled on demand. There will be an AI agent layer on every single PC out there. Since privacy will be an issue, "Shazam-like" filters will inhibit uncleared capture of voice. Makes sense?

A simple HTTPS, HTTP/3, SSL and security headers checker I built with AI 3 by dragonman | 1 comments on Hacker News. I’m a DevOps engineer and recently started experimenting seriously with AI-assisted coding to see how useful it actually is in real work. It checks: - HTTPS redirects - SSL certificate validity - Mixed content - basic security headers - HTTP/3 support AI helped a lot with speed — scaffolding, boilerplate, and quick iterations. But testing, edge cases, and reviewing security-related logic quickly reminded me that AI doesn’t replace understanding. You still own every line of code you ship. This is mainly a learning project, not meant to replace full security scanners. I’d appreciate any feedback, bug reports, or thoughts on what’s missing or misleading. Check it out: https://httpsornot.com