HNHacker News
TopNewBestAskShowJobs

danbrooks

262 karma · joined March 21, 2018

submissionscomments
danbrooks··on Statement from Dario Amodei on our discussions with the Department of War
Props to Dario and Anthropic for taking a moral stand. A rarity in tech these days.
danbrooks··on Tell HN: YC companies scrape GitHub activity, send spam emails to users
I got some emails like this from overseas developers looking to borrow my Linkedin to land a higher paying job.
danbrooks··on How not to answer the salary question
Interesting - I've never had that experience. In my experience the most variable part is the amount of stock/bonus on top of base salary.
danbrooks··on How not to answer the salary question
Yes. I pre-screen job postings to make sure that the compensation is workable before applying.
danbrooks··on X blames users for Grok-generated CSAM; no fixes announced
Could there be a criminal investigation? CSAM is a serious issue.
danbrooks··on Ask HN: What Are You Working On? (Nov 2025)
Built a web interface for Magic: the Gathering draft pick advice, trained on top player data.

Website: https://statisticaldrafting.com

Turns out to be a small niche, but I enjoy it!

danbrooks··on PSF has withdrawn $1.5M proposal to US Government grant program
I made a donation. Props to the PSF for standing up.
danbrooks··on I played 1k hands of online poker and built a web app with Cursor AI
I've heard (second-hand) that bots were instrumental in the decline of online poker popularity.
danbrooks··on The collapse of the econ PhD job market
Academic hiring has taken a big hit with the uncertainty and decline in federal funding.
danbrooks··on Show HN: Draw A Fish and watch it swim with the others:)
Surprisingly, this site is blocked under the "religion" category by my Fortune 50 employer. :)
danbrooks··on Ask HN: Why aren't AIs being used as app beta testers yet?
I worked with a team that did this for the Facebook app.

https://engineering.fb.com/2018/05/02/developer-tools/sapien...

danbrooks··on Making 2.5 Flash and 2.5 Pro GA, and introducing Gemini 2.5 Flash-Lite
This can happen if the prompt or response is blocked by a safety filter. Check some of the other fields in the response.
danbrooks··on As an experienced LLM user, I don't use generative LLMs often
As a data scientist, this mirrors my experience. Prompt engineering is surprisingly important for getting expected output - and use LLM POCs have quick turnaround times.
danbrooks··on Ask HN: What are you working on? (April 2025)
Building a Magic: the Gathering draft assistant trained on historical data.

Produces a pick order that shifts as the draft progresses.

Have a browser runnable colab notebook for 20+ sets.

https://github.com/danieljbrooks/statistical-drafting

danbrooks··on Tell HN: Announcing tomhow as a public moderator
Welcome Tom!
danbrooks··on Ask HN: How Do I Escape Homelessness After Rebuilding My Mental Health?
Sorry to hear about all that you are going through.

Looking at your profile - it appears as though you've worked extensively in technology in the past - and are reporting a cash crunch currently.

First order of business seems to be securing a place to live. Could be a friends couch or a room to rent.

Longer term, resuming employment in the tech industry seems to be a good path to financial security.

A couple of questions - is there a timeline for the inheritance money arriving? Will that enable you to rent a place to live? Do you plan on working concurrently with going back to school?

Again - sorry for all that you're going through - although it sounds like you have developed a positive attitude.

danbrooks··on Ask HN: SWEs how do you future-proof your career in light of LLMs?
Continual learning & keeping up with current trends as part of professional growth.

Working in ML my primary role is using new advances like LLMs to solve business problems.

It is incredible though, how quickly new tools and approaches turn over.

danbrooks··on My Ivy League school is threatening to suspend me if I don't shut down mystartup
Sorry to hear that. Consider consulting a lawyer to see what your options are.
danbrooks··on Show HN: I made a Mac app to search my images and videos locally with ML
Looks like image search using CLIP embeddings with frames sampled from video at a low frame rate.
danbrooks··on Ask HN: Best "entry tech" jobs to learn coding?
Breaking in can be tricky - but getting a position as an entry level developer sets you up on the software engineering/coding trajectory.
danbrooks··on 95%-ile isn't that good (2020)
When I was younger, I used to enjoy competition and climbing leaderboards. Now, parenting always comes first. Reaching the top 5% is possible for most people with sustained effort. Many participants aren't dedicated to reaching the top of the leaderboard.
danbrooks··on Fine-tuning Mistral 7B on Magic the Gathering Draft
Super interesting that drafts can be represented with LLMs.

The best performing draft AI's I've seen leverage representation learning in some form.

See: https://arxiv.org/pdf/2107.04438.pdf

danbrooks··on GraphCast: AI model for weather forecasting
Small changes in initial state can lead to huge changes down the line. See: the butterfly effect or chaos theory.

https://en.wikipedia.org/wiki/Chaos_theory

danbrooks··on How Bad Is Your Resume? – Get Your Resume Roasted by AI
Love the style, looks fun!
danbrooks··on Observation of zero resistance above 100 K in Pb₁₀₋ₓCuₓ(PO₄)₆O
Encouraging evidence that the recent work on room temperature superconductivity might be real!

Still waiting for confirmation from multiple labs.

danbrooks··on Can you simply brainwash an LLM?
Is this surprising? LLMs are trained to produce likely word/tokens in a dataset. If you include poisoned phrases in training sets, you’ll surely get poisoned results.
danbrooks··on Meta fined $1.3B over data transfers to U.S.
Is this comment GPT generated (following the description in meghan_rain's bio)?
danbrooks··on Big tech and the pursuit of AI dominance
I wouldn't say big tech was "trounced" by OpenAI. Advances in LLMs are driven by the research community. OpenAI is a small, well-resourced organization with the sole purpose of quickly deploying and hyping state of the art AI models. Big tech naturally has a slower product development cycle - and will make full use of LLMs in due time.
danbrooks··on TypeScripting the technical interview
The N queens puzzle is a well known mathematical problem.

https://en.wikipedia.org/wiki/Eight_queens_puzzle

danbrooks··on Ask HN: How did you rebuild yourself after having hit rock bottom?
I’m 5 years out from a tough last year of a PhD program. Married with kids now.

It gets better. Power through the final stretch.

← PreviousPage 2 of 5Next →