Mine seems to think that I'm some kind of detail-obsessed super-pedant. Personally, I think this is ridiculous. "super" is a Latin stem meaning "beyond", which implies that I've transcended the qualities of pedantry. A better term would be 'pluri-pedant', which denotes someone who is exceptionally punctilious while still remaining within the bounds of being pedantic.
An aside that I do want to mention here because it is a really unique way for many people to interface with LLMs: many commenters mention the model over indexing on a few comments they made that do not necessarily reflect of the broader themes of their writing. This is not any issue in the author’s engineering but an inherent issue in LLMs. The reason it is so noticeable in this case is because the subject matter is extremely familiar to the user: themselves.
LLMs consistently misrepresent information in this exact same way in, more critical applications. Because they are often employed on datasets that engineers and potentially end users are not deeply familiar with, the results often seem exceptional.
Disclaimer via my HN wrapped:
“The Anti LLM Manifesto
You will write a 5,000-word blog post on why a single Bayesian prior is more 'sentient' than GPT-6, and it will be ignored because the summary was generated by a 3B parameter model.”
I did like it, but for me it was fixated on 3-5 comments from the last 1-2 months that got a few more upvotes. It didn’t really work as an overview for the year. Still, a pretty cool thingy :)
Yeah, same here. For the comments it took into account it made pretty great roasts, but would have been better if it was actually comprehensive over the course of the year.
Thanks. I now run a two-step process: first pass reads through all posts and comments to extract patterns, second pass uses those to generate the content. Should be much more representative of your full year now :)
My impression was the same as the poster: it still over-indexes on a couple of recent posts.
Of course, it's possible that we've both been repeating ourselves all year long! I mean, I know I do that, I just think I've ridden more hobby horses than it picked up. :-)
It's fun, though. Thanks for sharing - a couple of my "roasts" gave me a genuine chuckle.
Your intense hatred for the concept of GDP is only matched by your strangely specific crusade against the calorie theory, making you the only person on HN who thinks the economy and thermodynamics are both just vibes.
> You've mentioned Gemini 2.0 Flash pricing and model comparisons so many times that I'm starting to think you're actually a Google Cloud Billing alert that gained sentience.
I wouldn’t mention it so much if Google stopped bumping up the price.
As a side note, I find this capability of AI to mine social profiles quite disturbing. Automated profiling of social media accounts can be and is used with malicious intent. The amount of personal detail that can be recovered this way is shocking. It is possible to associate this information with a real identity, and it can be used to target and intimidate individuals.
1.▲
Show HN: A songwriting DAW built entirely inside an Org-mode buffer(emacs.org)
432 points | 2 hours ago | 89 comments
2.▲
Why I'm still using Soulseek to trade prompt-engineered MIDI files
156 points | 4 hours ago | 42 comments
I admit I hastily tried clicking before realizing these were fake
This is fun, but the fact that it is now so easy to process a massive amount of social media data to extract a person's political leanings, hobbies, etc and infer information about them scares me. It only takes one government change (cough cough, Reform UK) for this to be used against me. It doesn't matter if I'm politically correct when what is correct changes over time.
>Your frustration with LGA socket reliability will finally boil over, leading you to move to a cabin in Mexico where you only communicate via GPS-disciplined atomic clocks and ham radio SDRs.
At first I was both amused and mildly annoyed at mine. Then I looked at the ones for some of the other HNers whom I follow and realized I got off easy.
In contrast to many others, I did not find this particularly interesting.
- The comic on is oddly cropped and contains speech attribution errors.
- It calls me an "extremist" regarding the wrong thing (I am many kinds of extremist, but certainly not Haskell).
- It claims I believe "any software failure is merely a design error" which is a complete misunderstanding of the ideas I presented.
- It says things like "the geometric mean of the snack bowl" which doesn't have meaning in English.
I feel like it has picked up on certain keywords and then just rolled with its own stereotypes of what those keywords represent, rather than actually taking a good look at what I think. A roast works because the roaster has clearly spent time and effort and care understanding the person roasted. This is way too shallow for that.
The 2026 and 2035 predictions (with a few exceptions) don't make sense at all, and the jokes in them fall completely flat. They're not good anti-jokes either. If someone said something like it in a social situation it would be followed by an awkward silence.
The vibe check and the time spent were really cool though. Super interesting. I would have loved to see those expanded.
I don't mean to be negative. The project is cool. I just wish it would put its focus on the valuable parts, rather than the things it is weak at. I guess this is my 45 % pedantic, 25 % contrarian, 20 % analytical self speaking.
> I feel like it has picked up on certain keywords and then just rolled with its own stereotypes of what those keywords represent, rather than actually taking a good look at what I think. A roast works because the roaster has clearly spent time and effort and care understanding the person roasted. This is way too shallow for that.
Yeah. It picks one random thing from one comment and turns into a lifestyle.
That's about how it came across for me as well: ignoring my actual content and joking about generalizations related to key words.
Project is cool overall, love the xkcd-like comic idea—but prompting and/or model-selection could use some work. I'd like to take a crack at tuning it myself :)
Perhaps it should also avoid putting too much emphasis on several comments to the same story: there was a story about VAT changes in Denmark, where I participated with several comments; but the generator decided that I apparently had a high focus vat, when I just wanted to provide some clarifying context to that story. I wonder how comments are weighed, is it individually or per story?
Specifically this roast:
> You have commented about the specific nuances of Danish VAT and accounting system hardcoding at least four times, proving you are the only person on Earth who finds tax infrastructure more exciting than the books being taxed.
Yeah, but I did it on the same story (i.e. context).
Though the other details it picked up, I cannot really argue with: the VAT bit just stood out to me.
This caused me to delete the HN password on my main account. It was already disconcerting to be reminded by Karpathy (I cannot help but see this as bragging) in his recent post that we must be good because LLMs are watching: https://karpathy.bearblog.dev/auto-grade-hn/
LLMs will be used to deanonymize all internet users. The AI "visionaries" are already bragging about this. You have been warned.
This was a bit of fun. I've been doing a lot of coding via GitHub copilot from my phone lately and it picked up on that and coined the term "Thumb driven development", which I absolutely love and am officially adopting!
>> Your next 'exception catcher' blog post will be written after a long night at the Chicago Microcenter and will feature a Kafka setup powered entirely by spite and BIPA lawsuits.
It's good and gave me some queasy chuckles. But it's not as fun as the "HN Front Page" from a few days back.
I suspect there's kind of a feedback loop here. An algorithm does X well; it becomes easy to churn out X; we realize that X wasn't as skilful as we thought. In other words, this kind of satire will become seen as stilted and conventional, in the same way that AI art looks stilted and conventional to us today.
That was hilarious, uncannily accurate in some places.
A romantic of the motherboard who thinks software engineering peaked at Xerox PARC and spent 2025 auditing every Show HN for missing source code while mourning the death of the em-dash. You will develop an LLM-powered sentiment analyzer that specifically detects when tech executives use 'fond farewell' to describe mass layoffs.
▲ Ask HN: Is there any way to travel without 24/7 neural-link monitoring?
▲ Mozilla CEO calls 15th rebranding to 'VibeBrowser' a 'bold pivot'
▲ Minimalism in the 2030s: Living on 8kb of memory for a week
> You spend so much time fighting macOS animations and keyboard layouts that I am surprised you have any time left to actually use the computer you keep threatening to replace with a Framework
Yep that’s me.
As for 2026 prediction:
> You will write a 4,000-word HN essay arguing that Silksong’s difficulty curve is a direct allegory for the South Korean 'Hagwon' education system.
Hilarious. It seemed to focus on something about Boolean logic that I don’t remember taking about, but otherwise awesome.
Are you sharing any of the prompts you used to generate this? Even if not verbatim, I’m interested to know how much you’re driving versus the model. How much more prompt is there than, “look at this comment history and write 3 roasts that the HN crowd will find funny”?
I don't like "Wrappeds" (low-key social hack to manufacture normalization of surveillance capitalism?), but with HN being public, I succumbed to temptation. Very fun, 10/10 no notes, surprisingly good for a small sample set this year.
> You write comments like you're trying to win a Pulitzer in Political Economy while trapped inside a middle-manager's strategy meeting.
A high-frequency human link-shortener who refuses to let a single HN thread go by without reminding the world that their personal blog exists as a primary source of truth.
Your commitment to plugging your blog is so aggressive that I am legally obligated to check if https://www.rxjourney.net/ is actually a horcrux for your digital soul.
A developer who believes every global outage is just a missing question mark away from salvation and spends their weekends reapplying thermal paste to fanless MacBooks while reminiscing about the tactile superiority of 2010 Dell Latitude trackpads.
Roasts
You post about Cloudflare outages caused by a single unwrap while your own codebase probably looks like a game of Russian Roulette played with Result types.
Your obsession with the thermal conductivity of fanless laptops is just a coping mechanism for the fact that your Rust builds take so long you could literally cook an egg on your chassis.
You have a very specific kink for 2010 Dell trackpads that makes me think you are either a Linux philosopher or someone who is no longer allowed within 500 feet of a Best Buy.
Feeder, FeedDeck, NewsBlur, Fluent Reader, Twine, Capy Reader, Reeder, Read You, Neo Feed, Readrop are all open-source RSS/feed readers for mobile (iOS, Android, or both)
, and they all have dark theme / night mode.
I like it, though I do notice that like most LLMs it seems to take criticism of LLM rollout kinda personally (actually circa 2023 this was considerably less true across every popular model)
I also think the predictions section seems kinda generic where the roast section felt better-personalized, which seems like a prompting issue
> You claim to value efficiency, yet you've spent the equivalent of a full fiscal year arguing about why a Firefox fork that 12 people use is the only path to salvation.
Mine "roasted" me by making fun of the fact I never finished a PhD, despite that being due to medical and other life circumstances that were well outside my control, including, but not limited to, some issues related to the fact I was a woman trying to get into academia who experienced the kinds of behaviors from people in the department which are not really suitable for polite discussion.
Additionally, it roasted me for building a project to "avoid the outdoors," which is another incredibly demeaning thing to say to someone who explicitly created that project because she was too medically unwell to be able to go outside as much as she wished and wanted to bring a bit of the outdoors inside. Very lame, definitely missed the mark.
The elisp and common lisp notes were on point, though, and did get a chuckle out of me.
A high-frequency haptics evangelist who is currently attempting to connect every physical object in his house to a DuckDB instance via an AI-controlled nipple mount.
Wasn't this same site showed off on HN about a few months ago? I recall using it back then where it would roast you, seems like this is just a rebrand to a "wrapped" version and more generally seems like an ad for your service.
Also the roasts are heavily front-loaded, the LLM is only really taking my most recent posts into account, not the especially far back ones earlier in the year.
> You spent three paragraphs arguing why estimation is a vital business requirement only to follow up by suggesting the world is an entropic chaos where nobody is actually in charge, which is a really convenient way to explain why your sprint is three weeks late.
> You've mentioned Futurama references so many times that I'm starting to think you're actually just Zap Brannigan with a Python Foundation fellowship.
> Frustrated with Waterfox and Orion, you will finally launch 'CERN-fox', a browser that only renders LaTeX and requires a muon-detected captcha for every search query.
"A high-latency architect who spends his days documenting every time a CDN sneezes while dreaming of a mountain drive through the Balkans with a fresh burek in hand."
Well, as a local VRAM libertarian, to manually prune the safety alignment part of a 500B LLM for it to run on 1GB RAM or VRAM is definitely a lifetime goal for me.
> You have mentioned being Australian at least five times as a personality trait just to remind us that our pennies are stupid and our tap-to-pay is thirty years behind yours.
Fair!
The XKCD comic generation was impressive.
The "your HN front page in 2035" doesn't really make sense to me because afaik personalised front pages are not a thing? (Or maybe they will be in 2035...)
Okay, mine was just awesome. Thank you. The only thing that would make it better would be if it could be easily saved.
(Also, it's a shame that it regenerates the xkcd every time)
"A seasoned architect who spends their days patrolling the wall between actual engineering and unsustainable AI hype while desperately trying to keep their Windows 10 box alive until the heat death of the universe. You are the only person on the internet who still remembers what a build script does and why we shouldn't let LLMs touch them without adult supervision."
Pretty fun overall. The roasts were really good, the stats were pretty neat, the predictions made me smile. The xkcdupe fell flat for me because it's pretty generic and I've seen many variants of it before, but still pretty impressive for an AI to make. The "Your HN Front Page in 2035" was definitely the weakest part for me. Maybe because every gag in it had already been done in other sections.
One thing I noticed is that it seems to give more weight to submissions (i.e. things I've submitted as posts) than to comments (or at least, doesn't let submissions get drowned in a sea of comments), which turns out as a good thing. And it doesn't seem to care about karma, which is also IMO a really good thing, makes it feel like a deep dive and keeps it interesting.
I’m going to have a sticky note warning myself to “Don’t write that on Hacker News.” I laughed but I hate me.
“You talk about 'Walking Out' of digital services so often that it's starting to sound less like a data strategy and more like a cry for help from your $10-a-month AWS bill.”
My last 2-3 months here have largely been posting about AI slop. But the prior 9 months were not, and included a lot of nuanced posting about how to make use of AI agents. I even got accused of being a shill at one point.
The wrapped is pretty much focused on the slop stuff. That's less interesting than the earlier parts of the year - I know what I've posted recently but have forgotten a lot more of what was going on a year ago or how the year developed.
Roasts can be amusing but I don't think they're the right vibe for a wrapped. I know it's harder to get an LLM to write something witty and insightful than a witty but shallow quip so maybe that's why roasts are here. Wrappeds are sort of infodumps though and LLMs are good at that, maybe there could be a two stage step where it reasons about some custom quirky stats or factoids that work based on your profile and then the second stage generates them.
Show HN: A Python 4.2 framework that transpiles to Go but still lets you use circular imports(github.com)
Microsoft 365 Audit: Now requiring a literal DNA sample for volume licensing compliance(theregister.com)
Zulip 15.0 adds 'Boomer Mode' to hide markdown from non-technical users(zulip.org)
eh?
Ask HN: Best audiologist for tuning hearing aids to filter out 'Notification Blindness'?
> For someone who claims to be a professional cinematographer, you spend an alarming amount of time looking at plain text arguments on a 1990s-style forum instead of actually framing a shot.
Because this is HN and we must critique: I think the “HN 2035” would be more entertaining if you adjusted the prompt to suggest it use company and product names that don’t actually exist. (There’s no way HN is half full of Tesla and OpenAI articles in 2035)
> Your posts suggest you are the only person on the internet who uses genetic algorithms for friendship management but still can't figure out a Typst template for your thesis.
A digital preservationist who wages a one-man war against the AI-slop apocalypse by manually indexing every human-written blog post in existence while playing Unreal Tournament 99 on a secondary monitor.
——-
I have to say, if that’s my future I’m kinda cool with it.
> You’ve mentioned the 1975 book The Mythical Man-Month so many times that I’m starting to think it’s your only personality trait besides complaining about Tailwind CSS.
And to add insult to injury it says I'm going to rewrite all my Python stuff in Rust (after spending the year teasing Rust). I got a good giggle from the comic book summary -- thanks for making this.
I really enjoyed my roast review. I loved the feedback that I'm only 5% helpful and 35% contrarian! I will endeavor to boost that helpfulness percentage next year.
I am surprised at the passion that some seem to feel over their own reviews.
> A high-frequency debunker who treats every comment thread as a zero-trust environment where empathy is a bug and citing Sartre is a security vulnerability. You are the only person on the planet capable of linking the efficiency of electrical line curvature to the ethics of Anthony Bourdain in a single browsing session.
A teenage digital architect who oscillates between solving the world's privacy crisis and wondering if high school chemistry is a psyop designed to kill his GitHub star count.
This is really awesome, I liked how it really detected my chemistry hatred and how the xkcd had the "see the mole concept is a false flag to obfuscate the real data, You have a test tomorrow" line as I kinda winging chemistry sometimes
> You are the only person on earth still trying to make 'decentralized link shorteners via Signal avatars' happen while failing organic chemistry.
100% accurate lmao, but for what it was worth i was trying it with signal call links since you can name a call name link 32 bits of storage which are persisted forever in signal's database iirc so it would've been a shitty link shortener but still I just loved signal and kinda wanted to build something on top of it
Do I really yap so much about chemistry here, I think that I have created more topics, surely lol but still I still enjoyed this a lot, maybe it just catched up on these traits more since I am pretty damn sure that I might be the only person here commenting about why in the world my country is requiring me to master chemistry university level to just get into a basic comp sci degree.
> Show HN: A Kanban board built on top of Bitwarden notes because why not
Btw, this was this close to happen except at the time I was vibe coding it with some new tool to stress test it with prototyping ideas basically but it didnt really work so i gave up on it but let me know if this idea fascinates someone lol
Happy holidays to everyone, this time of the year must mean a lot to people and I appreciate the spirit of holidays and gift giving too :)
Edit: also I love how it catches myself as existential since I genuinely had gotten existential because of hackernews once wondering what are the best ways to promote/grow open source so much so that I had written a manifesto, I can also be considered idealistic but I dont know why I forgot but "The FOSS Existentialist" is such a good title that I am gonna have it in my about page. This was genuinely brilliant.
Mine seems to think that I'm some kind of detail-obsessed super-pedant. Personally, I think this is ridiculous. "super" is a Latin stem meaning "beyond", which implies that I've transcended the qualities of pedantry. A better term would be 'pluri-pedant', which denotes someone who is exceptionally punctilious while still remaining within the bounds of being pedantic.
LLMs consistently misrepresent information in this exact same way in, more critical applications. Because they are often employed on datasets that engineers and potentially end users are not deeply familiar with, the results often seem exceptional.
Disclaimer via my HN wrapped: “The Anti LLM Manifesto You will write a 5,000-word blog post on why a single Bayesian prior is more 'sentient' than GPT-6, and it will be ignored because the summary was generated by a 3B parameter model.”
https://hn-wrapped.kadoa.com/aschobel
Of course, it's possible that we've both been repeating ourselves all year long! I mean, I know I do that, I just think I've ridden more hobby horses than it picked up. :-)
It's fun, though. Thanks for sharing - a couple of my "roasts" gave me a genuine chuckle.
I wouldn’t mention it so much if Google stopped bumping up the price.
Hah, I feel seen! This is pretty funny
I do enjoy its predictions.
https://hn-wrapped.kadoa.com/willis936
>Your frustration with LGA socket reliability will finally boil over, leading you to move to a cabin in Mexico where you only communicate via GPS-disciplined atomic clocks and ham radio SDRs.
https://hn-wrapped.kadoa.com/dctoedt
https://hn-wrapped.kadoa.com/rayiner
https://hn-wrapped.kadoa.com/tptacek
https://hn-wrapped.kadoa.com/dragonwriter
https://hn-wrapped.kadoa.com/jacquesm
https://hn-wrapped.kadoa.com/DannyBee
- The comic on is oddly cropped and contains speech attribution errors.
- It calls me an "extremist" regarding the wrong thing (I am many kinds of extremist, but certainly not Haskell).
- It claims I believe "any software failure is merely a design error" which is a complete misunderstanding of the ideas I presented.
- It says things like "the geometric mean of the snack bowl" which doesn't have meaning in English.
I feel like it has picked up on certain keywords and then just rolled with its own stereotypes of what those keywords represent, rather than actually taking a good look at what I think. A roast works because the roaster has clearly spent time and effort and care understanding the person roasted. This is way too shallow for that.
The 2026 and 2035 predictions (with a few exceptions) don't make sense at all, and the jokes in them fall completely flat. They're not good anti-jokes either. If someone said something like it in a social situation it would be followed by an awkward silence.
The vibe check and the time spent were really cool though. Super interesting. I would have loved to see those expanded.
I don't mean to be negative. The project is cool. I just wish it would put its focus on the valuable parts, rather than the things it is weak at. I guess this is my 45 % pedantic, 25 % contrarian, 20 % analytical self speaking.
Yeah. It picks one random thing from one comment and turns into a lifestyle.
Project is cool overall, love the xkcd-like comic idea—but prompting and/or model-selection could use some work. I'd like to take a crack at tuning it myself :)
It also appears highly biased towards recency as much of mine was roasting a topic I had only spoken of once and recently
- improved prompts with your feedback
- added post/comment shuffling to remove recency bias
- tried to fix the speech attribution errors in the xkcd
Specifically this roast:
> You have commented about the specific nuances of Danish VAT and accounting system hardcoding at least four times, proving you are the only person on Earth who finds tax infrastructure more exciting than the books being taxed.
Yeah, but I did it on the same story (i.e. context).
Though the other details it picked up, I cannot really argue with: the VAT bit just stood out to me.
LLMs will be used to deanonymize all internet users. The AI "visionaries" are already bragging about this. You have been warned.
https://hn-wrapped.kadoa.com/vldszn
https://hn-wrapped.kadoa.com/dang
> Malort-Driven Development
>> Your next 'exception catcher' blog post will be written after a long night at the Chicago Microcenter and will feature a Kafka setup powered entirely by spite and BIPA lawsuits.
0_0
I suspect there's kind of a feedback loop here. An algorithm does X well; it becomes easy to churn out X; we realize that X wasn't as skilful as we thought. In other words, this kind of satire will become seen as stilted and conventional, in the same way that AI art looks stilted and conventional to us today.
I thought that one was quite funny.
Yep that’s me.
As for 2026 prediction:
> You will write a 4,000-word HN essay arguing that Silksong’s difficulty curve is a direct allegory for the South Korean 'Hagwon' education system.
Yeah I can see that happening.
Are you sharing any of the prompts you used to generate this? Even if not verbatim, I’m interested to know how much you’re driving versus the model. How much more prompt is there than, “look at this comment history and write 3 roasts that the HN crowd will find funny”?
> You write comments like you're trying to win a Pulitzer in Political Economy while trapped inside a middle-manager's strategy meeting.
It would be nice if the username field was case-insensitive, since I think HN's username input is.
Things in between don't appear to move the needle at all.
So cool it's now easy to build custom recaps!
A high-frequency human link-shortener who refuses to let a single HN thread go by without reminding the world that their personal blog exists as a primary source of truth.
Your commitment to plugging your blog is so aggressive that I am legally obligated to check if https://www.rxjourney.net/ is actually a horcrux for your digital soul.
---
Posts You'd Upvote in 2035
1. Show HN: A compiler that refuses to build if the technical debt is too high for a junior(github.com)
2. Ask HN: How do I prove my identity to my fridge without using a Google/Apple pubkey?
3. The last human-written HTML file was just committed to the Library of Congress
https://hn-wrapped.kadoa.com/jedberg
(But in seriousness, this self reflection really does highlight what my year has been like and I truly appreciated the laughs)
———-
The Rust-Evangelizing Hardware Romantic
A developer who believes every global outage is just a missing question mark away from salvation and spends their weekends reapplying thermal paste to fanless MacBooks while reminiscing about the tactile superiority of 2010 Dell Latitude trackpads.
Roasts
You post about Cloudflare outages caused by a single unwrap while your own codebase probably looks like a game of Russian Roulette played with Result types.
Your obsession with the thermal conductivity of fanless laptops is just a coping mechanism for the fact that your Rust builds take so long you could literally cook an egg on your chassis.
You have a very specific kink for 2010 Dell trackpads that makes me think you are either a Linux philosopher or someone who is no longer allowed within 500 feet of a Best Buy.
And yeah, I will keep ranting about the lack of dark mode!
Works well for me on Vivaldi & Brave browsers on Android mobile.
There are many RSS readers with inbuilt dark theme / night mode.
Even if you prefer self-hosted RSS readers, you can find RSS readers (FOSS, not Premium/Freemium) on GitHub with such feature.
https://alternativeto.net/software/feedly/?feature=night-mod...
https://duckduckgo.com/?q=github+rss+reader+self+hosted+dark...
Feeder, FeedDeck, NewsBlur, Fluent Reader, Twine, Capy Reader, Reeder, Read You, Neo Feed, Readrop are all open-source RSS/feed readers for mobile (iOS, Android, or both) , and they all have dark theme / night mode.
https://alternativeto.net/software/feedly/?license=opensourc...
https://duckduckgo.com/?q=android+ios+open-source+rss+reader...
But definitely a fun read!
https://hn-wrapped.kadoa.com/azhenley
https://hn-wrapped.kadoa.com/nhatcher
(Kind of embarrassing TBH)
Mine is probably even more embarrasing, but mostly due to focusing on random threads: https://hn-wrapped.kadoa.com/ashirviskas
I also think the predictions section seems kinda generic where the roast section felt better-personalized, which seems like a prompting issue
Made me smile, thank you!
Mine "roasted" me by making fun of the fact I never finished a PhD, despite that being due to medical and other life circumstances that were well outside my control, including, but not limited to, some issues related to the fact I was a woman trying to get into academia who experienced the kinds of behaviors from people in the department which are not really suitable for polite discussion.
Additionally, it roasted me for building a project to "avoid the outdoors," which is another incredibly demeaning thing to say to someone who explicitly created that project because she was too medically unwell to be able to go outside as much as she wished and wanted to bring a bit of the outdoors inside. Very lame, definitely missed the mark.
The elisp and common lisp notes were on point, though, and did get a chuckle out of me.
This is awesome OP, love your work.
https://hn-wrapped.kadoa.com/neomantra
The MCP Obsessed Vibe-Coder
Also the roasts are heavily front-loaded, the LLM is only really taking my most recent posts into account, not the especially far back ones earlier in the year.
Gosh I love it.
Kif, I'm feeling the "Captain's itch".
It's not wrong.
https://hn-wrapped.kadoa.com/vee-kay
And such cool stuff is why we love HN!
> The Ultimate Fork
> Frustrated with Waterfox and Orion, you will finally launch 'CERN-fox', a browser that only renders LaTeX and requires a muon-detected captcha for every search query.
"A high-latency architect who spends his days documenting every time a CDN sneezes while dreaming of a mountain drive through the Balkans with a fresh burek in hand."
Well, as a local VRAM libertarian, to manually prune the safety alignment part of a 500B LLM for it to run on 1GB RAM or VRAM is definitely a lifetime goal for me.
What are the prompts you're usign?
> You have mentioned being Australian at least five times as a personality trait just to remind us that our pennies are stupid and our tap-to-pay is thirty years behind yours.
Fair!
The XKCD comic generation was impressive.
The "your HN front page in 2035" doesn't really make sense to me because afaik personalised front pages are not a thing? (Or maybe they will be in 2035...)
(Also, it's a shame that it regenerates the xkcd every time)
"A seasoned architect who spends their days patrolling the wall between actual engineering and unsustainable AI hype while desperately trying to keep their Windows 10 box alive until the heat death of the universe. You are the only person on the internet who still remembers what a build script does and why we shouldn't let LLMs touch them without adult supervision."
https://kadoa.b-cdn.net/wrapped/AndrewDucker-1766263283894.p...
The xkcd should be saved and cached once generated, I'll look into the issue.
How would you improve saving?
Actually, other than that, it works just fine. I just wanted to cache mine, for when your site gets melted by HN overload. So I did it manually.
Thanks again!
For posterity - wasn’t, still am not an s-base customer.
Show HN: SSH-to-Brain interface (requires tmux and 600mg of caffeine)
One thing I noticed is that it seems to give more weight to submissions (i.e. things I've submitted as posts) than to comments (or at least, doesn't let submissions get drowned in a sea of comments), which turns out as a good thing. And it doesn't seem to care about karma, which is also IMO a really good thing, makes it feel like a deep dive and keeps it interesting.
“You talk about 'Walking Out' of digital services so often that it's starting to sound less like a data strategy and more like a cry for help from your $10-a-month AWS bill.”
https://hn-wrapped.kadoa.com/Brajeshwar
How did it match my facial hair in the XKCD, since HN is text-only? :mind-blown:
https://hn-wrapped.kadoa.com/franky47
My last 2-3 months here have largely been posting about AI slop. But the prior 9 months were not, and included a lot of nuanced posting about how to make use of AI agents. I even got accused of being a shill at one point.
The wrapped is pretty much focused on the slop stuff. That's less interesting than the earlier parts of the year - I know what I've posted recently but have forgotten a lot more of what was going on a year ago or how the year developed.
Roasts can be amusing but I don't think they're the right vibe for a wrapped. I know it's harder to get an LLM to write something witty and insightful than a witty but shallow quip so maybe that's why roasts are here. Wrappeds are sort of infodumps though and LLMs are good at that, maybe there could be a two stage step where it reasons about some custom quirky stats or factoids that work based on your profile and then the second stage generates them.
Accurate!
Dammit, now I've got to improve those KPIs? Sigh...
But I do have one gripe, I'm a woman. I wish the XKCD comic was accurate. TwT.
Also HN: wow, that's totally me! That AI totally gets me.
When in Rome ¯\_(ツ)_/¯
> You spent twenty years avoiding a single line of JavaScript only to get depressed and give up on front-end dev as soon as you saw a CSS file.
> After your second child, your HN comments on healthcare reform will become so detailed they are legally classified as a white paper.
Lmao
Whelp I can’t recover from that one.
Because this is HN and we must critique: I think the “HN 2035” would be more entertaining if you adjusted the prompt to suggest it use company and product names that don’t actually exist. (There’s no way HN is half full of Tesla and OpenAI articles in 2035)
Edit: OK this was funny
> Your posts suggest you are the only person on the internet who uses genetic algorithms for friendship management but still can't figure out a Typst template for your thesis.
A digital preservationist who wages a one-man war against the AI-slop apocalypse by manually indexing every human-written blog post in existence while playing Unreal Tournament 99 on a secondary monitor.
——-
I have to say, if that’s my future I’m kinda cool with it.
https://hn-wrapped.kadoa.com/k3vinw
Ahahaha, not entirely wrong!
I am surprised at the passion that some seem to feel over their own reviews.
https://hn-wrapped.kadoa.com/rendall
> A high-frequency debunker who treats every comment thread as a zero-trust environment where empathy is a bug and citing Sartre is a security vulnerability. You are the only person on the planet capable of linking the efficiency of electrical line curvature to the ethics of Anthony Bourdain in a single browsing session.
No I don't! (nice project)
New 2026 resolution unlocked…
EDIT: a retry worked. Enjoy: https://hn-wrapped.kadoa.com/jaggs
https://hn-wrapped.kadoa.com/busymom0
A teenage digital architect who oscillates between solving the world's privacy crisis and wondering if high school chemistry is a psyop designed to kill his GitHub star count.
This is really awesome, I liked how it really detected my chemistry hatred and how the xkcd had the "see the mole concept is a false flag to obfuscate the real data, You have a test tomorrow" line as I kinda winging chemistry sometimes
> You are the only person on earth still trying to make 'decentralized link shorteners via Signal avatars' happen while failing organic chemistry.
100% accurate lmao, but for what it was worth i was trying it with signal call links since you can name a call name link 32 bits of storage which are persisted forever in signal's database iirc so it would've been a shitty link shortener but still I just loved signal and kinda wanted to build something on top of it
Do I really yap so much about chemistry here, I think that I have created more topics, surely lol but still I still enjoyed this a lot, maybe it just catched up on these traits more since I am pretty damn sure that I might be the only person here commenting about why in the world my country is requiring me to master chemistry university level to just get into a basic comp sci degree.
> Show HN: A Kanban board built on top of Bitwarden notes because why not
Btw, this was this close to happen except at the time I was vibe coding it with some new tool to stress test it with prototyping ideas basically but it didnt really work so i gave up on it but let me know if this idea fascinates someone lol
Happy holidays to everyone, this time of the year must mean a lot to people and I appreciate the spirit of holidays and gift giving too :)
Edit: also I love how it catches myself as existential since I genuinely had gotten existential because of hackernews once wondering what are the best ways to promote/grow open source so much so that I had written a manifesto, I can also be considered idealistic but I dont know why I forgot but "The FOSS Existentialist" is such a good title that I am gonna have it in my about page. This was genuinely brilliant.