Hacker News: thesandlord

New comment by thesandlord in "Ask HN: How do you use 5–10 minute gaps productively?"

thesandlord — Mon, 05 Jan 2026 00:38:52 +0000

I know everyone here is telling you not to hyper optimize, but I've found drafting a prompt for an AI agent (Cursor/Devin/Codex/etc) and letting it run and then using the remaining time to do a quick body weight set (pushups, squats)

New comment by thesandlord in "Sick of smart TVs? Here are your best options"

thesandlord — Sat, 13 Dec 2025 00:49:25 +0000

New Hisense TVs have USB-C DisplayPort support. Pretty cool, but realistically I don't see how it's different from HDMI from a usefulness standpoint.

Edit: It is cool I can plug my phone or laptop into the TV with one cable, no adapters, and get some power as well. For some reason it didn't work with my Steam Deck which was strange.

New comment by thesandlord in "Nokia N900 Necromancy"

thesandlord — Fri, 12 Dec 2025 03:23:19 +0000

I remember running around campus looking for WiFi hotspots with my N810, using Google Voice to text my friends ($0.10 per text, no thank you!). Learned so much Linux admin skills that became so useful later in life. Favorite device ever! Eventually moved to Android smartphones but the ease of hacking, the amazing community (internettablettalk.com, looks like its gone now :( unfortunately...)

New comment by thesandlord in "Show HN: Inkeep (YC W23) – Agent Builder to create agents in code or visually"

thesandlord — Thu, 16 Oct 2025 14:05:07 +0000

Is this primarily for building chat based agents? What if I want to trigger a workflow via API or webhook and the wait for some sort of human in the loop verification? Do you have an example for something like that?

The visual UI + code is really cool, addresses the weaknesses of both approaches.

New comment by thesandlord in "Using a laptop as an HDMI monitor for an SBC"

thesandlord — Thu, 09 Oct 2025 20:40:17 +0000

I played all of KOTOR2 on Xbox using a capture card without sound. Good times.

New comment by thesandlord in "Using a laptop as an HDMI monitor for an SBC"

thesandlord — Thu, 09 Oct 2025 15:21:13 +0000

If you have an iPad with a USB-C port, you can use the free Orion app to do this too

https://orion.tube/

YC Startups Use AI: Agents, OCR, and Prompt Engineering with Mercoa (YC W23)

thesandlord — Wed, 06 Aug 2025 16:29:40 +0000

Article URL: https://www.aiengineering.report/p/how-yc-startups-use-ai-agents-ocr

Comments URL: https://news.ycombinator.com/item?id=44814240

Points: 4

# Comments: 0

New comment by thesandlord in "Use the Gemini API with OpenAI Fallback in TypeScript"

thesandlord — Mon, 07 Apr 2025 04:30:48 +0000

I've been using using [BAML](https://github.com/boundaryml/baml) to do this, and it works really well. Lets you have multiple different fallback and retry policies, and returns strongly typed outputs from LLMs.

New comment by thesandlord in "Splitting engineering teams into defense and offense"

thesandlord — Mon, 14 Oct 2024 22:02:53 +0000

Don't do that! This was a great post with a lot to learn from.

The fact you came to a very similar solution from first principles is very interesting (assuming you didn't know about this before!)

New comment by thesandlord in "Invoice OCR with type-safe LLM calls"

thesandlord — Tue, 13 Aug 2024 16:01:39 +0000

CTO here, using BAML in our stack has let us use LLMs a LOT more in places that wouldn't make sense without the type-safety. We started using it for line item tag prediction and classification, false-positive checking on our fuzzy match algo, and forecasting. Especially with models like 4o-mini and Gemini Flash, its now actually cost effective to do this at scale.

New comment by thesandlord in "Why technical cofounders reject you"

thesandlord — Mon, 22 Jul 2024 15:27:44 +0000

What incentive would HN or YC have to do this?

Especially since they are putting out content that argues that you don't even need a business co-founder: https://youtu.be/43RhhwpiSk0?si=NMt8jLFLB9dU0TQ0

New comment by thesandlord in "Self-driving Waymos secure final clearance for expansion beyond S.F"

thesandlord — Fri, 21 Jun 2024 22:17:42 +0000

In my experience, after tip Lyft/Uber are more expensive. The exception is during surges when Waymo's price shoots up like crazy.

New comment by thesandlord in "Veo"

thesandlord — Tue, 14 May 2024 21:43:11 +0000

We use GPT-4o for data extraction from documents, its really good. I published a small library that does a lot of the document conversion and output parsing: https://npmjs.com/package/llm-document-ocr

For straight OCR, it does work really well but at the end of the day its still not 100%

New comment by thesandlord in "Show HN: Kubernates in Node.js"

thesandlord — Mon, 06 May 2024 18:17:52 +0000

https://news.ycombinator.com/item?id=26661223

New comment by thesandlord in "Why You Should Be a Luddite"

thesandlord — Thu, 01 Feb 2024 05:41:00 +0000

I think a portion of people would say yes, it's better to keep the status quo. Imo progress always wins because eventually someone somewhere will break the taboo. Technology even gets lost and rediscovered, which means even if you supress the knowledge someone will still eventually do it. So it's pretty pointless to fight, I'd rather try to benefit from it.

New comment by thesandlord in "My First Week at Google (2005)"

thesandlord — Thu, 18 Jan 2024 19:42:30 +0000

I left Google in 2020. It is not like this anymore. I heard its gotten even worse.

New comment by thesandlord in "Ask HN: Help me solve a dilemma, study in the US or take a job in the EU?"

thesandlord — Mon, 15 Jan 2024 23:35:54 +0000

> Also there is the fact that America is extremely anti-immigration at the moment.

In what way? Compared to which countries? If OP is coming to the USA for school, I don't think this is a concern esp in the short term.

Show HN: Document OCR for Node.js with GPT Vision

thesandlord — Thu, 16 Nov 2023 17:59:50 +0000

GPT Vision is probably the best generic OCR and document parsing tool out there. It can handle invoices, contracts, receipts, and even hand written notes. However, there are some limitations to the API that make it hard to work with:

1) No support for JSON output 2) No support for PDFs

llm-document-ocr is a simple Node library that does these pre and post processing steps for you. It converts PDFs into PNGs, crops whitespace around the images, and parses the JSON output.

Hope this saves you some time if you are building your own OCR stack on top of GPT and other LLMs!

Comments URL: https://news.ycombinator.com/item?id=38293007

Points: 1

# Comments: 1

New comment by thesandlord in "Expanded Silverado EV Production Delayed a Year as GM Cites 'Evolving EV Demand'"

thesandlord — Tue, 24 Oct 2023 00:05:22 +0000

And they wanted to cancel the Bolt too...

New comment by thesandlord in "Tesla announces Cybertruck deliveries in November, with 125k production capacity"

thesandlord — Wed, 18 Oct 2023 21:05:04 +0000

Anecdote: I saw a Cybertruck driving on a highway in SF last week. Didn't get a great look but it was definitely real!