What are you working on?  Any new ideas which you&#x27;re thinking about?

Ask HN: What Are You Working On? (June 2025)

First of all, this is purely a personal learning project for me, aiming to combine three of my passions: photography, software engineering, and my family memories. I have a large collection of family photos and want to build an interactive experience to explore them, ala Google or Apple Photo features.My goal is to create a system with smart search capabilities, and one of the most important requirements is that it must run entirely on my local hardware. Privacy is key, but the main driver is the challenge and joy of building it myself (an obviously learn).The key features I&#x27;m aiming for are:Automatic identification and tagging of family members (local face recognition).Generation of descriptive captions for each photo.Natural language search (e.g., &quot;Show me photos of us at the beach in Luquillo from last summer&quot;).I&#x27;ve already prompted AI tools for a high-level project plan, and they provided a solid blueprint (eg, Ollama with LLaVA, a vector DB like ChromaDB, you know it). Now, I&#x27;m highly interested in the real-world human experience. I&#x27;m looking for advice, learning stories, and the little details that only come from building something similar.What tools, models, and best practices would you recommend for a project like this in 2025? Specifically, I&#x27;m curious about combining structured metadata (EXIF), face recognition data, and semantic vector search into a single, cohesive application.Any and all advice would be deeply appreciated. Thanks!

Ask HN: What's the 2025 stack for a self-hosted photo library with local AI?

In the early 80&#x27;s in the US, a popular DIY electronics magazine had a book of the month club that I loved. Most were small and leather bound hardback with topics like: make your own hydrophone; augmented reality (required a full room and a boom arm, sadly); an LCD model rocket launcher ignition; computer vision; lots and lots of robots.One book I remember (large, softcover, yellow cover) featured black and white, pen and ink illustrations of fantastically complex robots and machines. One that I remember was a water-based machine with video camera eye mounted on a tripod of pontoons. Wow, these illustrations filled my dreams.Does anyone remember this? Do you remember the name of the illustrator? Anything at all?

Ask HN: 80s electronics book club; anyone remember this illustrator?

My kids are not quite screen time age, but at some point will be. I&#x27;d like to give them an interesting computer experience instead of just plopping them in front of an iPad with some media.When I was a kid I had fond memories of exploring the file system, figuring out how applications worked, playing with Kid Pix, Paint, and a few games (roughly Apple IIGS, Macintosh 2, through iMac + a Windows XP desktop).Do you have any fun old laptops or device you&#x27;ve got lying around that you&#x27;ve used to introduce kids into a desktop environment?Any and all recommendations welcome :)

Ask HN: How have you shared computers with your young child (~3 to 5)

There is a lot of debate whether AI will surpass humans in all economically viable skills (AGI, by one definition). Regardless of whether this will happen, or when, many people already have lost their jobs in part due to the emerging capabilities of AI models, including writing, document analysis, design, art, etc.This leaves many in a position where they fear they will be next on the chopping block. Many assume physical tasks will take longer since it will take longer to build up, verify and test humanoid robots vs. some virtual AI agent. However, many believe the writing is on the wall either way, and those in domains involving using their hands or bodies will only have a few more years than the formerly employed white-collar class.Which skills then, or combinations of skills, do you believe will be safest for staying employed and useful if AI continues improving at the rate it has been for the past few years?

Ask HN: Which skill do you believe will take the longest to be replaced by AI?

I asked this question before about cellphones, now I&#x27;m in the market for a tablet.Is there an Android device that comes without pre-installed bloatware, gets long-term support, and stays up to date with the latest Android versions?

Ask HN: Stock Android tablet free of bloatware?

Noticed today while throwing out a cartridge that the dye sublimation used leaves the negatives in each color (blue, yellow, magenta, white) for each photo and the details are enough to reconstruct the pictures if someone spent a little time to automate via Photoshop or gimp to layer and merge the inverse of each color. Searching for this on Google didn&#x27;t seem to produce any relevant results so I figured I&#x27;d post here to be corrected on my assumptions, or at least raise awareness that it&#x27;s a concern. Canon publishes quite a bit of support docs around privacy for the digital usage side of devices, but I guess old school exfiltrarion is still a possibility.Anyone at Canon or relevant industries have any insight? What&#x27;s a proper and privacy friendly consumer disposal option for these cartridges?

Canon selphy cp1500 privacy concerns

It seems like black on white text is now truly verboten for whatever reason on at least 70-80% of websites.can anyone explain the benefits? to me, everything looks faded out, and rather than my eyes being able to pick up the letterforms I just perceive a field of gray and have to rely on plug-ins to get easily discernible text back.

Ask HN: How did low contrast text become so pervasive?

(this was posted last year, and i thought it&#x27;s time to repeat it. i mostly copied the text and rules to avoid confusion. i hope that&#x27;s ok)Let&#x27;s match Free Software and Open Source projects that need help with people looking to contribute. Think of this as &quot;Who&#x27;s Hiring&quot; but for Free Software or Open Source projects looking for volunteers - a semi regular thread to surface interesting projects that could use more hands.Please include: Project name and description (if not widely known); Tech stack; Areas needing help (DOCS, CODE, DESIGN, etc.); Level (BEGINNER-FRIENDLY if applicable); Email address, Project website or other means of contacting you.Ground rules:Post only if you maintain&#x2F;run the projectOne post per project&#x2F;suiteNo commercial recruitmentNo thread complaintsReaders: Only reach out if you actually want to contribute.

Ask HN: Which Free Software or Open Source Project Needs Help?

telnet:&#x2F;&#x2F;issue3.anewsession.com

A literary magazine accessible only via telnet

Simple question. Where and how do you host your Go apps? I feel like either you have to run a VM or pay for the complexity of a Google cloud.I&#x27;m sure some people will now say things like Fly or Railway but curious to know firsthand.Personally I&#x27;m still using DigitalOcean, I git pull, compile from source and run the Go binary, occassionally with a shell script. It&#x27;s fronted by nginx and certbot&#x2F;letsencrypt. That&#x27;s it. For some reason I wish this was some simple solution instead of the endless variety of hosting out there. I always worry about Fly, Railway or someone else going out of business. I find other tools really complicated, and dedicated app hosting too expensive. A VM plus some open source works well. But I guess when you offloading that hosting to someone else you start expecting all sorts of tools. Maybe if there was just a dedicated CLI based thing. Who knows.

Ask HN: Where do you host your Go apps

Any good new apps or revamped old apps?Im currently using Safari on iPhone. It works.

Ask HN: How Are You Reading HN in June 2025?

Since Google Glass made its debut in 2012, there&#x27;s been a fair amount of hype around augmented reality and related tech coming into its own in industry, presumably enhancing worker productivity and capabilities.But I&#x27;ve heard and seen so little use in any industries. I would have thought at a minimum that having access to hands-free information retrieval (e.g. blueprints, instructions, notes, etc), video chat and calls for point-of-view sharing, etc would be quite useful for a number of industries. There do seem to be interesting pilot trials involving Hololens in US defense (IVAS) as well as healthcare telemonitoring in Serbia.Do you know of any relevant examples or use cases, or are you a user yourself? What do you think are the hurdles - actual usefulness, display quality, cost, something else?

Ask HN: Anyone using augmented reality, VR, glasses, helmets etc. in industry?

I use it all the time, so this must be recent.

Tell HN: (dictionary|thesaurus).reference.com is now a spam site

I&#x27;m noticing that the HN header clips quite close to the rest of the content on my phone as of today. Is anyone else noticing this or is it a bug on my end?Edit to add: the delete comment page has the header moved to the footer.

Ask HN: Is the header CSS broken for you?

I&#x27;m a big fan of the &quot;noprocrast&quot; setting on HN, but lately it seems like it&#x27;s no longer functioning, allowing for seemingly infinite procrastination.Anyone else seeing this issue?

Ask HN: Is noprocrast still working for you?

After 5 years of building and fighting for our startup, we’ve reached the end — the product will be shut down soon. I won’t mention names to keep this from sounding promotional. Let’s just say it’s a kind of website builder.We’ve tried (unsuccessfully) to sell the codebase. Meanwhile, some of our most loyal users are now asking us to open source it. Part of me feels this would be a meaningful way to give back and ensure the project doesn’t completely disappear.However, I can also foresee a lot of technical and legal complications, not to mention potential maintenance burdens.Has anyone here been through this before? Any lessons, regrets, or advice?Thanks a lot in advance!(AI used to improve spelling)

Ask HN: Startup shutting down, should we open source?

Hey all - I really love emacs as a text editor and environment, but I&#x27;m wondering...
Does there exist something that is- More customizable than emacs- more mature community and &#x27;giants to stand on&#x27; (like packages)- Stronger expected lifespan- Realistic enough to actually commit to (not like, build the universe, build your OS from scratch)I&#x27;d love to hear if there&#x27;s anything you&#x27;ve found beyond the Mariana Trench!
Will

Something 'deeper' than Emacs, or am I looking for a unicorn?

What do you use for private service monitoring in your organization, and how did you configure it initially?

Ask HN: What do use for private service monitoring?

Hi everyone. I&#x27;m on a hunt to find all languages that are designed specifically to compile to WASM. I have a project (hram.dev -- hand-rolled assembly machine) that I want to build to share the joy of unwrapping a new computer in the 80s&#x2F;90s that boots up with an editor so that you can program it directly in assembly, and I plan to use wamr+llvm for near-native performance while still having isolation so that you can mess things up. Obviously the ability to write WAT directly will be fundamental and certainly fun, but I am looking for higher level languages that make it slightly less convenient to write, to bundle with it internally so that users have at least two choices of how to write code. Do you know of any other languages designed specifically for wasm? These are all I could find:Most likely:curlywas (https:&#x2F;&#x2F;github.com&#x2F;exoticorn&#x2F;curlywas) -- c-like but very low-level; seems very complete; rust impl; mit license; short but seemingly thorough docs?wa (https:&#x2F;&#x2F;github.com&#x2F;wa-lang&#x2F;wa) -- go-like, not clear how high&#x2F;low level it is, thorough docs, seems promising, agpl license, lots of mandarin in docsvirgil (https:&#x2F;&#x2F;github.com&#x2F;titzer&#x2F;virgil) -- ruby-like? gc; cant find license; last commit 3 hours ago; thorough docs but all in md files in repoassemblyscript (https:&#x2F;&#x2F;github.com&#x2F;AssemblyScript&#x2F;assemblyscript) -- typescript-like; apache 2 license, minimal runtime with gc; implemented in jswalt (https:&#x2F;&#x2F;github.com&#x2F;ballercat&#x2F;walt) -- JavaScript-like made for wasm, 25 contributors! decent looking docs, might actually be usable! most commits 7 years ago but last commit 3 years ago though, mit licenseonyx (https:&#x2F;&#x2F;wasmer.io&#x2F;posts&#x2F;onyxlang-powered-by-wasmer, https:&#x2F;&#x2F;github.com&#x2F;onyx-lang&#x2F;onyx) -- ocaml-like? recent activity, full docs, bsd license, not sure if it has lower level capabilities or how much the higher level features cost at runtime or build timewaforth (https:&#x2F;&#x2F;github.com&#x2F;remko&#x2F;waforth) -- forth for wasm! upside is that its forth, downside is that its forth; mit license; great docs; seemingly inefficient due to constant lookups?Less likely:thinscript (https:&#x2F;&#x2F;github.com&#x2F;evanw&#x2F;thinscript) -- js-like with macros; abandoned 9 years ago; not yet licensedwase (https:&#x2F;&#x2F;github.com&#x2F;area9innovation&#x2F;wase) -- C-like syntax but still wasm-like, not super recent but not super old, only a few contributors, not sure how complete it is but its docs give a feeling of being mostly-complete, MIT licensewam (https:&#x2F;&#x2F;github.com&#x2F;kanaka&#x2F;wam) -- wasm macro preprocessor, just one guy, last commit 7 years ago, very few built in macros, mozilla license (???)wah (https:&#x2F;&#x2F;github.com&#x2F;tmcw&#x2F;wah) -- wasm but with infix, doesn&#x27;t seem extensible with macros, two contributors, last commit 8 years ago, eclipse license (???)Honorable mentions:mini-c (https:&#x2F;&#x2F;github.com&#x2F;maierfelix&#x2F;mini-c) -- C to wasm compiler, seemingly abandoned 8 years ago, not sure how complete it isc4wa (https:&#x2F;&#x2F;github.com&#x2F;kign&#x2F;c4wa) -- c to wasm compiler, no activity in 3 years, no license, written in java

Ask HN: Languages Designed for WASM?

I recently came up with a backup strategy that seems so simple I assume it must already exist — but I haven’t seen it in any mainstream tools.The idea is:The latest backup (timestamped) always contains a full copy of the current source state.Any previous backups are stored as deltas: files that were deleted or modified compared to the next (newer) version.There are no version numbers — just timestamps. New versions can be inserted naturally.Each time you back up:1. Compare the current source with the latest backup.2. For files that changed or were deleted: move them into a new delta folder (timestamped).3. For new&#x2F;changed files: copy them into the latest snapshot folder (only as needed).4. Optionally rotate old deltas to keep history manageable.This means:The latest backup is always a usable full snapshot (fast restore).Previous versions can be reconstructed by applying reverse deltas.If the source is intact, the system self-heals: corrupted backups are replaced on the next run.Only one full copy is needed, like a versioned rsync mirror.As time goes by, losing old versions is low-impact.It&#x27;s user friendly since the latest backup can be browsed through with regular file explorers.Example:Initial backup:latest&#x2F;
 a.txt # &quot;Hello&quot;
 b.txt # &quot;World&quot;Next day, a.txt is changed and b.txt is deleted:latest&#x2F;
 a.txt # &quot;Hi&quot;
backup-2024-06-27T14:00:00&#x2F;
 a.txt # &quot;Hello&quot;
 b.txt # &quot;World&quot;The newest version is always in latest&#x2F;, and previous versions can be reconstructed by applying the deltas in reverse.I&#x27;m curious: has this been done before under another name? Are there edge cases I’m overlooking that make it impractical in real-world tools?Would love your thoughts.

A reverse-delta backup strategy – obvious idea or bad idea?

One of the greatest hackers, wrote allot of foundational nodejs modules.

Ask HN: What Happened to James Halliday ( Substack)?

What's the best gem you've found on Hacker News?

I&#x27;m interested to know if there&#x27;s a pattern in the cars purchased by the HN community (programmers, lawyers, C-suite people, etc.)

Ask HN: What's Your Car?

Hi, I know many of you here are running B2B start-ups. What authentication service&#x2F;framework are you using for your multi-tenant authentication?I am the CEO of Backtick (backtickai.com) - an AI-native issue tracker designed to remove manual task management. Our product is built using mostly open-source frameworks and products; we strongly adhere to this ideology in our company. However, we currently use PropelAuth, which is not open-source, so we are looking to migrate away from it.Better-auth seems like a good option. It supports multi-tenancy out of the box, has good documentation, and is MIT licensed. However, it&#x27;s very new, not battle-tested, and I&#x27;m unsure about its security vulnerabilities.Next-auth, while battle-tested, makes setting up multi-tenancy a pain.If it&#x27;s between these two, my preference would be Next-auth because of its maturity. But I&#x27;m wondering how you all are handling authentication. Which framework or hosted service are you using?I would appreciate your comments on this.

Ask HN: Better-auth or Nextauth or something else

For example, why don&#x27;t beta testing services such as TestFlight have ChatGPT as a possible beta tester along with the human testers?

Ask HN: Why aren't AIs being used as app beta testers yet?

I’m hosting a turn-based multiplayer browser game on a single Hetzner CCX23 x86 cloud server (4 vCPU, 16GB RAM, 80GB disk). The backend is built with Node.js and Socket.IO and is run via Docker Swarm. I use also use Traefik for load balancing.Matchmaking uses a round-robin sharding approach: each room is always handled by the same backend instance, letting me keep game state in memory and scale horizontally without Redis.Here’s the issue:
At ~500 concurrent players across ~60 rooms (max 8 players&#x2F;room), I see low CPU usage but high event loop lag. One feature in my game is typing during a player&#x27;s turn - each throttled keystroke is broadcast to the other players in real-time. If I remove this logic, I can handle 1000+ players without issue.Scaling out backend instances on my single-server doesn&#x27;t help. I expected less load per backend instance to help, but I still hit the same limit around 500 players. This suggests to me that the bottleneck isn’t CPU or app logic, but something deeper in the stack. But I’m not sure what.Some server metrics at 500 players:- CPU: 25% per core (according to htop)- PPS: ~3000 in &#x2F; ~3000 out- Bandwidth: ~100KBps in &#x2F; ~800KBps outCould 500 concurrent players just be a realistic upper bound for my single-server setup, or is something misconfigured? I know scaling out with new servers should fix the issue, but I wanted to check in with the internet first to see if I&#x27;m missing anything. I’m new to multiplayer architecture so any insight would be greatly appreciated.

Ask HN: Why does my Node.js multiplayer game lag at 500 players with low CPU?

Hi all, my Meta developer account was suddenly restricted without warning, and there’s no way to appeal or reach support. I believe this may have been triggered by a location (read, IP address) change: I’m a student-entrepreneur who recently moved from China to Kenya for school holidays.I’m building a WhatsApp-based accounting tool for African small businesses, and this restriction is blocking critical operations. I’ve come across others facing the same issue, with no review option or clear reason:
<a href="https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;WhatsappBusinessAPI&#x2F;comments&#x2F;1kvolvn&#x2F;facebook_developer_account_restricted_no_review&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;WhatsappBusinessAPI&#x2F;comments&#x2F;1kvolv...</a><a href="https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;facebook&#x2F;comments&#x2F;1h1c79w&#x2F;developer_account_suspended_no_warning_no&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;facebook&#x2F;comments&#x2F;1h1c79w&#x2F;developer...</a>Some of the people in these posts were able to contact Meta support, but it seems like Meta has disabled direct support chats.Thanks in advance.

Tell HN: Meta developer account suspended

been using cloudflare for a while. was wondering what others are using as an alternative

Ask HN: Alternatives to Cloudflare for DNS?

Hey everyone,I’ve been thinking a lot about webhook delivery reliability lately. In many projects I’ve worked on, building robust webhook infra turned out to be deceptively complex:- Retry logic (exponential backoff, timeouts)
- Handling non-2xx responses
- Delivery monitoring and alerting
- Back-pressure or queueing to avoid overwhelming receivers
- Secure signing and validation flowsIn one project, a failed webhook caused a payment processing delay for hours because the retry logic was buggy. Another time, burst traffic took down the receiver endpoint with no DLQ strategy in place.I’ve been researching different approaches teams here use:Do you build your own custom webhook delivery queue and monitoring system?
 Use cloud solutions like AWS EventBridge or Step Functions to orchestrate?
 Or integrate third-party tools that handle delivery, retries, and observability for you?I’m curious about how you ensure production-grade reliability at scale without burning dev hours on plumbing. Recently, I’ve been working on a tool in this space to handle these issues automatically, but would love to hear:- What architecture have you found most reliable?
- What are the edge cases you’ve encountered (e.g. signature mismatches, downstream outages)?
- Any horror stories or lessons learned from webhook failures in production?Looking forward to learning from your experiences and best practices around webhook infra!

How do you handle production webhook delivery reliability in your apps?

If couples find that parenting isn&#x27;t for them (e.g., within the first year of their baby&#x27;s life), they would be able to place the baby for adoption easily and without stigma.Would this encourage more couples to have children?

Ask HN: Would this idea help address declining populations in many countries?

Hello everyone,Conventional wisdom on productivity suggests that the closer we get to our goals, the more motivated we become. The shrinking distance to the finish line should, logically, be our most potent fuel.However, I&#x27;ve recently been analyzing a deeply counter-intuitive pattern—an anomaly I&#x27;ve started calling &quot;The 90% Gravity.&quot;The pattern is this:There&#x27;s a statistically significant &quot;Danger Zone,&quot; not at the start of a project, but in the final stretch—roughly between 80% and 95% completion.In this zone, rates of procrastination, self-sabotage, and near-abandonment spike disproportionately. It&#x27;s as if a palpable, invisible force actively repels us from the very success we are about to grasp. This isn&#x27;t just fatigue; the pattern holds even for an individual&#x27;s most desired, passion-fueled projects. In fact, the more meaningful the goal, the stronger the pull of this negative gravity seems to be.If this pattern holds true, it suggests our greatest adversary isn&#x27;t the inertia of starting, but a strange form of &#x27;success aversion&#x27; that ambushes us when victory is already in sight.I wanted to open this up to the community here:1. Have you personally experienced this &quot;90% Gravity&quot;? A project you were passionate about, only to inexplicably stall when it was almost done?
2. Theoretically, what psychological forces do you believe are at play here? Is it a fear of the success itself? A fear of the void that comes after a long-held goal is achieved? Something else entirely?I&#x27;m curious to read your perspectives.

The 90% Gravity Problem: Why We Tend to Quit Right Before the Finish Line

Hi HN,As the title suggested, any recommendations and tips to build in LLM assistants with Vim? Any experiences to share about such a setup?I’ve only dabbled with AI-assisted coding using Gemini &#x2F; ChatGPT “on the side”, rather than a fully integrated experience like Cursor.

Ask HN: LLM Assisted Vim Workflows?

Yes, I know I could just ask ChatGPT, but I still believe in good old HN wisdom, and I’m also curious about the entrepreneurial point of view. It is surprising to see there has been attempts for everything in the past, and this community always knows something about it.

Ask HN: Is it possible to generate usable energy from environmental heat?

Ask HN: Will MCP replace GUI interacting with back end via RESTful APIs?

Hey folks — I&#x27;m looking to hand off a project I&#x27;ve been working on: Eyeball, a clean and minimal bookmarking app for iOS that uses AI to help users save, organize, and understand their links.Website: https:&#x2F;&#x2F;eyeball.wtf&#x2F;
App: https:&#x2F;&#x2F;apps.apple.com&#x2F;us&#x2F;app&#x2F;eyeball-ai-bookmarks-notes&#x2F;id6670705634The app is live on the App Store, has several hundred users.Selling because my founding engineer took a full-time role and I don’t have the bandwidth to grow it solo. Would love to see it in the hands of someone who wants to take it further. Thinking a simple profit-share if it ever generates revenue?If you’re into calm productivity tools, AI&#x2F;notes, or second-brain apps, this could be a fun one to take on.If you’re interested, drop your contact details and I&#x27;ll reach out.