<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: graerg</title><link>https://news.ycombinator.com/user?id=graerg</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 17 Nov 2026 02:08:37 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=graerg" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by graerg in "Ask HN: What are you working on? (November 2026)"]]></title><description><![CDATA[
<p>You and me both :(</p>
]]></description><pubDate>Mon, 09 Nov 2026 15:47:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=50427179</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=50427179</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50427179</guid></item><item><title><![CDATA[New comment by graerg in "Ask HN: What are you working on? (November 2026)"]]></title><description><![CDATA[
<p>I used to really enjoy watching Dota 2 tournaments. I also like Bake-Off, and Battle-Bots was a great idea.<p>I'm tinkering on a project to take two contestants and dump them into a reasonably nicely equipped linux environment (checkpointed fly.io Sprites), giving them a task, and watching them compete in real time to see who gets it done first. All of it streams to twitch.<p>Here's an example match (I have two agents pretending to be contestants for the sake of the demo: <a href="https://www.twitch.tv/videos/2923688013" rel="nofollow">https://www.twitch.tv/videos/2923688013</a>). Basically anything goes; I imagine people will clone+stow their dotfiles, leverage their workflows/harnesses/editor configs, etc. Viewers will get to see how its done.<p>Next on the todo list is "weight classes" that let contestants compete with a fixed level of harness capability: natty means no agents at all, small local open weight LLM tier only (qwen, gemma, etc), large open weight models only (deepseek, glm, kimi), frontier lite (haiku, luna), frontier (6.1-sol, haiku), frontier plus (astra, fable).<p>Ultimately I want to rank users in each weight class so people can compete for delivering the most at various cost tiers.</p>
]]></description><pubDate>Mon, 09 Nov 2026 01:41:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=50419911</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=50419911</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50419911</guid></item><item><title><![CDATA[New comment by graerg in "Rubin Observatory's first LSST Camera release: 500k galaxies in the COSMOS field"]]></title><description><![CDATA[
<p>Hey! Former astronomer here; I used to work a lot with SDSS data products. Do you happen to have a link to any sort of photometry DB/catalog for Rubin? Basically I’m looking to grab a ton of data but I’ve been out of the field for a while and I’m not sure where they publish their data sets and which ones you’d recommend as being relatively accessible.</p>
]]></description><pubDate>Wed, 05 Aug 2026 19:23:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=49187672</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=49187672</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49187672</guid></item><item><title><![CDATA[New comment by graerg in "Show HN: LastShelf – an emergency map of your family's documents bills& contacts"]]></title><description><![CDATA[
<p>I built if-i-go-missing.com along these lines. Weird, I’m also a Brown!</p>
]]></description><pubDate>Thu, 09 Jul 2026 16:06:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=48848197</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48848197</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48848197</guid></item><item><title><![CDATA[New comment by graerg in "Ask HN: Who wants to be hired? (July 2026)"]]></title><description><![CDATA[
<p>Location: Orange County, CA (Pacific Time) 
Remote: Yes (preferred); hybrid Orange County also fine 
Willing to relocate: No 
Technologies: Python, Go, SQL | FastAPI, Django, Vue, HTMX | Postgres, TimescaleDB, DuckDB, Redis | Temporal, Airflow, NATS | AWS (S3/RDS/EC2/Redshift), Docker, Kubernetes | Prometheus/Grafana | LLMs in production (RAG, eval, orchestration, structured extraction) 
Resume/CV: <a href="https://brojonat.com/Jon_Brown_Resume.pdf" rel="nofollow">https://brojonat.com/Jon_Brown_Resume.pdf</a> 
Website / writing: <a href="https://brojonat.com" rel="nofollow">https://brojonat.com</a> 
GitHub: <a href="https://github.com/brojonat" rel="nofollow">https://github.com/brojonat</a> 
Email: brojonat@gmail.com<p>PhD-trained engineer/manager (astronomy by training -- data is data). Currently Senior Data Science Manager at Hyundai Motor America's North American Safety Office, where I build large-scale vehicle telemetry pipelines, LLM-driven classification and RAG over enterprise data, and led department-wide LLM/AI adoption from technical workshops to production SDLC.<p>Previously managed a data science team at HeadSpin: built data science pipelines, killed error-prone spreadsheet workflows with proper APIs and dashboards, reimplemented the flagship data product to reduce costs, latency, and churn.<p>Looking for senior data / platform sorts of roles. Equally happy IC or manager -- I just want to ship. Open to full-time or fractional/contract; happy to even work on small projects you want to send my way to help get to know each other.<p>I build a lot on my own time, and the side projects are probably the best read on how I work:<p>- forohtoo -- Go service for awaiting Solana payments via SSE/NATS, with a clean `Await(memo, amount)` SDK. Writeup: <a href="https://brojonat.com/posts/forohtoo/" rel="nofollow">https://brojonat.com/posts/forohtoo/</a> 
- IncentivizeThis (<a href="https://incentivizethis.com" rel="nofollow">https://incentivizethis.com</a>) -- bounty-based authentic ad platform using LLM-as-judge across Reddit, YouTube, Twitch, Instagram, HN, TripAdvisor. Writeup: <a href="https://brojonat.com/posts/incentivizethis/" rel="nofollow">https://brojonat.com/posts/incentivizethis/</a> 
- If I Go Missing (<a href="https://if-i-go-missing.com" rel="nofollow">https://if-i-go-missing.com</a>) -- dead-man's-switch service (Go + Temporal + Postgres + Twilio + Stripe). 
- IYBI ("If You Build It [They Will Come]", <a href="https://iybi-twc.com" rel="nofollow">https://iybi-twc.com</a>) -- websites for independent workers, managed entirely by email. Customer pays via Stripe, emails an AI agent (Claude + Cloudflare Email Worker + R2 + K8s CronJob), the agent researches them, builds the site, and handles ongoing updates. My own consulting site runs on it.<p>Mostly Go or Python services running on my personal k8s cluster. Happy to talk shop or small projects we could work on together to assess fit.</p>
]]></description><pubDate>Wed, 01 Jul 2026 17:29:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=48750378</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48750378</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48750378</guid></item><item><title><![CDATA[New comment by graerg in "Ask HN: What are you working on? (June 2026)"]]></title><description><![CDATA[
<p>For my demos I've been running 3 minute sessions but I'm planning to run 30-60 minute sessions depending on the scenario. I want to see people push the boundary of what can be done in an hour (with an agent or without, it's up to the contestant!) and ultimately have the match VODs serve as entertainment but also reference examples for "good" development workflows.</p>
]]></description><pubDate>Sun, 14 Jun 2026 18:42:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=48531000</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48531000</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48531000</guid></item><item><title><![CDATA[New comment by graerg in "Ask HN: What are you working on? (June 2026)"]]></title><description><![CDATA[
<p>I'm working on a competitive coding gameshow. I'm imagining a combination of great british bakeoff, battle bots, and dota. Basically contestants get dropped into a fully equipped dev machine (all the bells and whistles one could want/expect including neovim, agent harnesses, cool styling, etc and if you want you can always clone your dotfiles and stow them!). I've gotten a decent prototype that live streams from Fly.io sprites to twitch, and I'm able to voice over or have OpenAI do commentary on the match. I've got a demo here: <a href="https://www.twitch.tv/videos/2792893261" rel="nofollow">https://www.twitch.tv/videos/2792893261</a>. Still a ways to go, but it seemed like a fun way to tinker with Sprites.</p>
]]></description><pubDate>Sun, 14 Jun 2026 18:07:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=48530559</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48530559</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48530559</guid></item><item><title><![CDATA[New comment by graerg in "Expanding Project Glasswing"]]></title><description><![CDATA[
<p>It is already very plausible (and has been since the 1950s) without the advent of LLMs. This is just another layer on top of the preexisting and very plausible existential threats we already face.</p>
]]></description><pubDate>Wed, 03 Jun 2026 02:21:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=48379112</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48379112</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48379112</guid></item><item><title><![CDATA[New comment by graerg in "Ask HN: Who wants to be hired? (June 2026)"]]></title><description><![CDATA[
<p>Location: Orange County, CA (Pacific Time)
Remote: Yes (preferred); hybrid Orange County also fine
Willing to relocate: No
Technologies: Python, Go, SQL | FastAPI, Django, Vue, HTMX | Postgres, TimescaleDB, DuckDB, Redis | Temporal, Airflow, NATS | AWS (S3/RDS/EC2/Redshift), Docker, Kubernetes | Prometheus/Grafana | LLMs in production (RAG, eval, orchestration, structured extraction)
Resume/CV: <a href="https://brojonat.com/Jon_Brown_Resume.pdf" rel="nofollow">https://brojonat.com/Jon_Brown_Resume.pdf</a>
Website / writing: <a href="https://brojonat.com" rel="nofollow">https://brojonat.com</a>
GitHub: <a href="https://github.com/brojonat" rel="nofollow">https://github.com/brojonat</a>
Email: brojonat@gmail.com<p>PhD-trained engineer/manager (astronomy by training -- data is data). Currently Senior Data Science Manager at Hyundai Motor America's North American Safety Office, where I build large-scale vehicle telemetry pipelines, LLM-driven classification and RAG over enterprise data, and led department-wide LLM/AI adoption from technical workshops to production SDLC. Previously managed a data science team at HeadSpin: built data science pipelines, killed error-prone spreadsheet workflows with proper APIs and dashboards, reimplemented the flagship data product to reduce costs, latency, and churn.<p>Looking for senior data / platform sorts of roles. Equally happy IC or manager -- I just want to ship. Open to full-time or fractional/contract; happy to even work on small projects you want to send my way to help get to know each other.<p>I build a lot on my own time, and the side projects are probably the best read on how I work:<p>- forohtoo -- Go service for awaiting Solana payments via SSE/NATS, with a clean `Await(memo, amount)` SDK. Writeup: <a href="https://brojonat.com/posts/forohtoo/" rel="nofollow">https://brojonat.com/posts/forohtoo/</a>
- IncentivizeThis (<a href="https://incentivizethis.com" rel="nofollow">https://incentivizethis.com</a>) -- bounty-based authentic ad platform using LLM-as-judge across Reddit, YouTube, Twitch, Instagram, HN, TripAdvisor. Writeup: <a href="https://brojonat.com/posts/incentivizethis/" rel="nofollow">https://brojonat.com/posts/incentivizethis/</a>
- If I Go Missing (<a href="https://if-i-go-missing.com" rel="nofollow">https://if-i-go-missing.com</a>) -- dead-man's-switch service (Go + Temporal + Postgres + Twilio + Stripe).
- IYBI ("If You Build It [They Will Come]", <a href="https://iybi-twc.com" rel="nofollow">https://iybi-twc.com</a>) -- websites for independent workers, managed entirely by email. Customer pays via Stripe, emails an AI agent (Claude + Cloudflare Email Worker + R2 + K8s CronJob), the agent researches them, builds the site, and handles ongoing updates. My own consulting site runs on it.<p>Mostly Go or Python services running on my personal k8s cluster. Happy to talk shop or small projects we could work on together to assess fit.</p>
]]></description><pubDate>Mon, 01 Jun 2026 21:46:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48363102</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48363102</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48363102</guid></item><item><title><![CDATA[New comment by graerg in "SQLite is all you need for durable workflows"]]></title><description><![CDATA[
<p>That's the whole thesis; YAGNI.</p>
]]></description><pubDate>Sat, 30 May 2026 14:41:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=48336766</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48336766</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48336766</guid></item><item><title><![CDATA[New comment by graerg in "SQLite is all you need for durable workflows"]]></title><description><![CDATA[
<p>I run my own temporal service in my k8s cluster; this setup is the backbone for almost all my applications. For simplicity, I opted for the postgres backend. You still need to run the 4 (?) other service (history, matching, frontend, ui, maybe others, definitely others if you want observability with prometheus/grafana, and tad bit more complexity if you want tailscale to get in there and poke around).<p>They ship Helm charts so reality is somewhere between "helm deploy" and "substantial ops burden". I don't have to touch it very frequently, but that is not to say I don't have to touch it. There's occasional releases and there have been times where (probably due to my inexperience with helm) I botched an upgrade and lost some data. And I've been on this journey for years; when I first started, they didn't have a Python SDK and it was one of my (many) excuses to learn Go. But anyway to your point, yes, if you're comfortable with k8s and Helm then you shouldn't have much of a problem running hundreds of thousands of workflows; if you want to really push the throughput and optimize cost you probably need to get creative the individual services and look into cassandra (maybe? idk).</p>
]]></description><pubDate>Sat, 30 May 2026 14:36:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48336693</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=48336693</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48336693</guid></item><item><title><![CDATA[New comment by graerg in "Tell HN: OpenAI silently removed Study Mode from ChatGPT"]]></title><description><![CDATA[
<p>They made a big point of explicitly advertising this as a feature with the GPT-5 rollout, no? Routing to cheaper models/less reasoning depending on the input prompt.</p>
]]></description><pubDate>Sun, 12 Apr 2026 17:19:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=47742146</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=47742146</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47742146</guid></item><item><title><![CDATA[Ask HN: Supplemental Consulting Work Advice]]></title><description><![CDATA[
<p>I have a full time job. I'm considering taking on additional consulting work (ideally in some sort of fractional manner), but I'm a little suspicious about the companies that have seemed open to such an arrangement. What are some best practices to follow in order to come out of this sort of thing successful and not scammed? I'd imagine "work with people you know" is the first step, but I'm curious if others have any lessons they've learned so I don't have to learn them the hard way. Some friends of a friend (totally different industry than mine) were recently fired for pursuing additional work and it's spurred some paranoia.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47542319">https://news.ycombinator.com/item?id=47542319</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 27 Mar 2026 13:18:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=47542319</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=47542319</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47542319</guid></item><item><title><![CDATA[New comment by graerg in "The only moat left is knowing things"]]></title><description><![CDATA[
<p>Exactly! It must be exhausting to have this huge preoccupation with determining if something has come from an LLM or not. Just judge the content on it's own merits! Just because an LLM was involved doesn't mean the underlying ideas are devoid of value. Conversely, the fact that an LLM wasn't involved doesn't mean the content is worth your time of day. It's annoying to read AI slop, but if you're spending more effort suspiciously squinting at it for LLM sign versus assessing the content itself, then you're doing yourself a disservice IMO.</p>
]]></description><pubDate>Thu, 29 Jan 2026 12:47:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=46809468</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=46809468</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46809468</guid></item><item><title><![CDATA[New comment by graerg in "Turso is an in-process SQL database, compatible with SQLite"]]></title><description><![CDATA[
<p>You could make the case uv falls in this category (I just prefix all my pip commands with uv) though we have yet to see if astral will become a "successful business"; I'm hoping they pull it off.</p>
]]></description><pubDate>Fri, 23 Jan 2026 03:27:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=46728026</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=46728026</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46728026</guid></item><item><title><![CDATA[New comment by graerg in "Operating Margins"]]></title><description><![CDATA[
<p>Part of it is this tufte.css: <a href="https://edwardtufte.github.io/tufte-css/" rel="nofollow">https://edwardtufte.github.io/tufte-css/</a></p>
]]></description><pubDate>Fri, 14 Nov 2025 14:05:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=45926782</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=45926782</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45926782</guid></item><item><title><![CDATA[New comment by graerg in "I’ve removed Disqus. It was making my blog worse"]]></title><description><![CDATA[
<p>There was a post on HN a few weeks ago about natalie.sh replacing blog comments with Bluesky threads. <a href="https://news.ycombinator.com/item?id=44826164">https://news.ycombinator.com/item?id=44826164</a><p>I was inspired by that and wrote up my experience integrating with my Hugo based blog here: <a href="https://brojonat.com/posts/bluesky-hugo-comments/" rel="nofollow">https://brojonat.com/posts/bluesky-hugo-comments/</a></p>
]]></description><pubDate>Tue, 30 Sep 2025 14:42:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=45426107</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=45426107</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45426107</guid></item><item><title><![CDATA[New comment by graerg in "Study mode"]]></title><description><![CDATA[
<p>People keep citing this study (and it was on the top of HN for a day). But this claim falls flat when you find out that the test subjects had effectively no experience with  LLM equipped editors and the 1-2 people in the study that actually <i>did</i> have experience with these tools showed a marked increase in productivity.<p>Like yeah, if you’ve only ever used an axe you probably don’t know the first thing about how to use a chainsaw, but if you know how to use a chainsaw you’re wiping the floor with the axe wielders. Wholeheartedly agree with the rest of your comment; even if you’re slow you lap everyone sitting on the couch.</p>
]]></description><pubDate>Tue, 29 Jul 2025 19:19:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=44727257</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=44727257</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44727257</guid></item><item><title><![CDATA[New comment by graerg in "Show HN: I'm giving away $100 on IncentivizeThis"]]></title><description><![CDATA[
<p>I should have said “traditional ads” but yes you’re right. Product placement is one of many use cases. You could also sponsor bounties with more general requirements in order to drive engagement in a particular sector, or incentivize comments with a particular sentiment (e.g., I have a bounty for encouraging comments on hobbyist subreddits).</p>
]]></description><pubDate>Wed, 16 Jul 2025 14:53:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=44582997</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=44582997</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44582997</guid></item><item><title><![CDATA[Show HN: I'm giving away $100 on IncentivizeThis]]></title><description><![CDATA[
<p>I built a platform that lets businesses incentivize online discussion via bounties. Instead of ads, businesses post bounties for specific content and metrics they want to see borne out (e.g., an engaging Reddit comment that mentioning my business and has at least 100 upvotes). Then, the denizens of the internet rise to the challenge and redeem the bounty for a reward.<p>To demonstrate how it works, I'm putting my money where my mouth is. This post links to $100 in bounties ($10 each) for posting an engaging comment on this thread. Limit one bounty per person (and per comment).<p>I modeled the full bounty lifecycle with Temporal workflows, so they're completely durable. Bounty fulfillment (i.e., content verification) is done with """AI""", and payouts are done with USDC on Solana so they're really fast. Happy to discuss this more in the comments.<p>For more info, check out incentivizethis.com; there's a one-minute demo video on the landing page that shows the gist. There's also a longer (5 minute) onboarding video here which should show you how to get up and running as a bounty hunter: <a href="https://finicky-elephant.clueso.site/share/aef8d2a6-b07b-4409-9ab5-920985b56404" rel="nofollow">https://finicky-elephant.clueso.site/share/aef8d2a6-b07b-440...</a>.<p>Finally, if you think it's neat and want to be kept in the loop, make sure you join the Discord; that's where I'll be coordinating with bounty hunters as I roll out more capabilities. Here's the link if you missed it: <a href="https://discord.gg/uvzgf6r2" rel="nofollow">https://discord.gg/uvzgf6r2</a>.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44582030">https://news.ycombinator.com/item?id=44582030</a></p>
<p>Points: 2</p>
<p># Comments: 4</p>
]]></description><pubDate>Wed, 16 Jul 2025 13:21:09 +0000</pubDate><link>https://www.incentivizethis.com/bounties/bounty-dec88db5-3d13-4042-95b5-e681b428afe5</link><dc:creator>graerg</dc:creator><comments>https://news.ycombinator.com/item?id=44582030</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44582030</guid></item></channel></rss>