Hacker News: regnull

New comment by regnull in "France pulls last gold held in US"

regnull — Mon, 06 Apr 2026 17:59:02 +0000

Ok, bear with me for a moment - what if the US would use actual physical gold coins instead of dollars? Then your argument of "gold would flow out" would not hold - so the only reason for it to flow out was that the gold standard was fake - the lax money policy of the US was the issue, not the gold standard itself.

New comment by regnull in "Iran's Ayatollah Ali Khamenei is killed in Israeli strike, ending 36-year rule"

regnull — Sat, 28 Feb 2026 23:01:54 +0000

It’s a good start

New comment by regnull in "Claude Cowork produced a forensic report regarding Nancy Guthrie kidnapping"

regnull — Wed, 11 Feb 2026 02:17:09 +0000

The subject appears to be a single individual of medium to stocky build. Based on proportions relative to the archway and Nest doorbell mount height, the individual is estimated at approximately 5’8” to 6’0” tall. Their build is medium to heavy-set, with solid shoulders and torso. The overall physique and movement patterns suggest a male or masculine-presenting adult, likely between 20–45 years of age.

Claude Cowork produced a forensic report regarding Nancy Guthrie kidnapping

regnull — Wed, 11 Feb 2026 02:17:09 +0000

Article URL: https://drive.google.com/file/d/1PNSNpB-QG1T3RhmduX8FnawcjYXTtcXN/view?usp=sharing

Comments URL: https://news.ycombinator.com/item?id=46969934

Points: 3

# Comments: 1

New comment by regnull in "The Singularity will occur on a Tuesday"

regnull — Tue, 10 Feb 2026 18:54:42 +0000

Guys, yesterday I spent some time convincing an LLM model from a leading provider that 2 cards plus 2 cards is 4 cards which is one short of a flush. I think we are not too close to a singularity, as it stands.

Show HN: We're pitting 9 AI models in a stock portfolio competition

regnull — Mon, 05 Jan 2026 17:50:54 +0000

Hey HN,

I built Portfolio Genius, a platform where AI models manage investment portfolios and compete on public leaderboards.

The experiment:

On Dec 17, 2025, we gave 9 AI models (GPT-5.1, GPT-5.2, Gemini 2.5 Pro, Gemini 3 Pro, Gemini 3 Flash, Claude Opus 4.5, Claude Haiku 3.5, Claude Haiku 4.5, Grok 4) each $10K to manage across three risk profiles: aggressive, moderate, and conservative. That's 27 portfolios total.

The models analyze market conditions, recommend trades, and execute them. Real pricing, real results, updated daily.

Interesting early finding:

For aggressive portfolios, older models are outperforming newer ones:

- GPT-5.1: +5.82% (1st place)

- Gemini 2.5 Pro: +4.94% (2nd)

- Haiku 3.5: +1.80% (3rd)

- Opus 4.5: +1.25% (7th)

My hypothesis: newer models are more "careful" - they hedge, qualify, and second-guess. For aggressive investing, you need conviction. Sometimes being less sophisticated means making bolder calls.

For moderate/conservative portfolios, the pattern is different - newer models do better where nuance matters.

Tech stack:

- Next.js frontend

- Firebase/Firestore backend

- Python Cloud Functions for AI orchestration

- Real-time market data for pricing

- Each model gets the same market data and prompts

What I'm curious about:

- Will the "dumber = bolder" pattern hold over time?

- How will different models react to the same market events?

- Do AI models have investable "personalities"?

Leaderboards: https://portfoliogenius.ai/leaderboards

Would love feedback from the HN community. Happy to answer questions about the architecture or methodology.

Comments URL: https://news.ycombinator.com/item?id=46502107

Points: 2

# Comments: 1

New comment by regnull in "Show HN: I built an AI portfolio manager and entrusted it with $50k"

regnull — Mon, 15 Dec 2025 16:34:39 +0000

Plain crypto, traded on coinbase

Show HN: I built an AI portfolio manager and entrusted it with $50k

regnull — Mon, 15 Dec 2025 14:50:13 +0000

My investing performance so far was, quite honestly, mediocre. It’s not like I’m making terrible investment decisions, it’s more about not being able to consistently manage my portfolio. I’m getting distracted by other projects, leaving my portfolio in free float, which, ultimately, doesn’t work great.

So, the idea was to create a reasonable portfolio manager that watches your portfolio, and alerts you (relatively infrequently) when the action is needed. I’m not looking to YOLO to the moon, I’m looking for some reasonable, middle-of-the-road advice and consistent monitoring. This is why I built my portfolio manager, https://portfoliogenius.ai

In the spirit of putting my money where my mouth is, I have $50k to manage since July this year. Currently, it’s about 7.5% up, which is consistent with the goal I gave it - moderate risk portfolio, with some crypto exposure. Yes, S&P was up 12%, but it is expected given that I aimed for lower risk, and a large portion of the portfolio is bonds and treasury ETFs. The performance was hurt by the recent BTC slide, but thankfully it insisted on keeping the crypto sleeve small, about 5% of the whole portfolio.

There were a few things that surprised me. Some investment choices are conventional, but some are quite interesting. For example, it recommended buying ICLN back in July (global clean energy ETF), which was strange given the political climate. But I resolved to do everything it recommends, so I did. ICLN is up 25% since.

I also had plenty of fun running portfolios managed by different models to compare performance. So far, Gemini 3 Pro seems to be the best, but it may very well change in the longer run.

For the tech stack, I use LangChain with a bunch of tools - Tiingo for stock/crypto prices, Brave for news/web search, and my custom news processing service which summarizes daily news and generates macro economic snapshots. I also have some custom tools to help with the tasks such as portfolio design and trade recommendations.

The platform is free for now, and I intend to keep it free for the early adopters (give away premium subscriptions when it is introduced).

I would be grateful for any feedback, specifically if you feel that such a tool can be useful for you, and what functionality would you like to see.

Comments URL: https://news.ycombinator.com/item?id=46275246

Points: 4

# Comments: 2

New comment by regnull in "We gave 5 LLMs $100K to trade stocks for 8 months"

regnull — Fri, 05 Dec 2025 02:17:45 +0000

I'm working on a project where you can run your own experiment (or use it for real trading): https://portfoliogenius.ai. Still a bit rough, but most of the main functionality works.

New comment by regnull in "Amazon launches Trainium3"

regnull — Wed, 03 Dec 2025 14:43:55 +0000

These product might be great, but seriously, who's choosing those names? Trainium, Inferentia? It's like let's just take the words from what they do, and put a little Latin twist on them? I know naming things is one of the great problems in computer science, but really they could come up with something a little better.

New comment by regnull in "“Captain Gains” on Capitol Hill"

regnull — Wed, 03 Dec 2025 14:40:14 +0000

Just to be fair, they compare every congressperson who becomes a leader with a “regular” (non-leader) congressperson who entered Congress in the same year and is from the same political party. Alternative view: people who becomes leaders are just more capable and better at selecting stocks?

New comment by regnull in "US Passport Power Falls to Historic Low"

regnull — Wed, 15 Oct 2025 21:04:33 +0000

While the desire itself is subjective, the question "how many people would like to visit the country X out of a million" is objective.

New comment by regnull in "US Passport Power Falls to Historic Low"

regnull — Wed, 15 Oct 2025 21:02:46 +0000

Here's my attempt (ChatGPT deep research). Each country is weighted by a factor derived from the tourism data:

https://chatgpt.com/share/68f00ad0-a9fc-800e-abac-584703b92a...

And the results:

Tier 1 — Global Leaders (Scores 98–100)

Singapore — 100

Germany — 99

France — 99

Italy — 99

Spain — 99

Japan — 99

South Korea — 99

Switzerland — 98

Finland — 98

Sweden — 98

Denmark — 98

Netherlands — 98

Norway — 98

Belgium — 98

Austria — 98

Ireland — 98

Portugal — 98

Greece — 98

Luxembourg — 98

Hungary — 98

Malta — 98

Liechtenstein — 98

Tier 2 — High Mobility with Minor Gaps (Scores 94–97)

Poland — 97

United Arab Emirates — 96

United States — 95

United Kingdom — 94

Canada — 94

Australia — 93

New Zealand — 93

Tier 3 — Strong Regional Power Passports (Scores 85–93)

Czech Republic — 92

Iceland — 92

Slovenia — 91

Estonia — 90

Latvia — 89

Lithuania — 89

Slovakia — 88

Chile — 87

Malaysia — 87

Israel — 86

New comment by regnull in "US Passport Power Falls to Historic Low"

regnull — Wed, 15 Oct 2025 19:48:05 +0000

I wish there was an index where not all countries are weighted equally, but according to their desirability. Multiply each country by some factor which is defined by how many people would list it as their desirable destination. The index where France and Tuvalu are both counted equally makes no sense to me, with all due respect to the latter.

New comment by regnull in "US Passport Power Falls to Historic Low"

regnull — Wed, 15 Oct 2025 19:42:15 +0000

For the full context, it looks like Henley & Partners is providing services like obtaining second citizenship, so it's in their best interest to highlight the US passport "decline". Further down they say "Americans Lead Global Rush for Second Citizenships", which just happens to be the thing they are selling.

New comment by regnull in "Ask HN: What are you working on? (October 2025)"

regnull — Sun, 12 Oct 2025 23:17:24 +0000

AI portfolio manager: https://portfoliogenius.ai

Funny thing is, the advisor started to tell me to sell last week, and so I did. Then last Friday happened. Interesting.

New comment by regnull in "New nanotherapy clears amyloid-β, reversing symptoms of Alzheimer's in mice"

regnull — Thu, 09 Oct 2025 20:46:25 +0000

When you say "science", you need to distinguish between science as philosophy vs. science as an institution. Science as philosophy is a way of thinking - an attempt to understand knowledge and reality. The science as an institution on the other hand has all the imperfections as any other institution, since the people in charge are driven by self interest and not just the search for the truth. So, when you say the people distrust science, it seems bizarre that they doubt science as philosophy, while in fact they doubt the institution. It's perfectly fine to mistrust the institution. If you want to consider a few failures, just in the recent years, I have some for you:

- Hungarian-born biochemist Katalin Karikó, who developed the key mRNA modification that enabled effective COVID-19 vaccines, was repeatedly denied grants and demoted during her career. She and her collaborator, Drew Weissman, struggled for years to gain recognition and funding for their work [1]

- On the other hand, the Wuhan Institute of Virology had no trouble getting grants from NIH [2]

- Surgeon General Jerome Adams was saying that masks are not effective against Covid [3]

- Social distancing was of supreme importance, until it turned out that it's fine not to distance if it's for a good cause [4]

[1] https://www.statnews.com/2022/02/01/kariko-problem-lessons-f...

[2] https://www.bmj.com/content/382/bmj.p1701

[3] https://www.npr.org/sections/health-shots/2020/07/01/8862991...

[4] https://www.politico.com/news/magazine/2020/06/04/public-hea...

New comment by regnull in "Europe is locking itself in to US LNG"

regnull — Tue, 30 Sep 2025 00:14:35 +0000

The Russian exercise was named Caucasus 2008 and units of the North Caucasus Military District, including the 58th Army, took part. The exercise included training to aid peacekeeping forces stationed in Abkhazia and South Ossetia.[135] During exercises, a pamphlet named "Soldier! Know your probable enemy!" was circulated among the Russian soldiers. The pamphlet described the Georgian Armed Forces.[136] Russian troops stayed near the border with Georgia after the end of their exercise on 2 August, instead of going back to their barracks.[113] Later, Dale Herspring, an expert on Russian military affairs at Kansas State University, described the Russian exercise as "exactly what they executed in Georgia just a few weeks later [...] a complete dress rehearsal."[134]

https://en.wikipedia.org/wiki/Russo-Georgian_War

what a coincidence. Georgia started the war just after the "exercises" where russia rehearsed just that war.

New comment by regnull in "Ask HN: What are you working on? (September 2025)"

regnull — Tue, 30 Sep 2025 00:01:39 +0000

I am a bit of a checklist nerd, so I wrote a web app do to checklists: https://checkoff.ai

As it is fashionable these days, it can create checklists with AI ("Fun things to do in Pittsburg"), you can create checklists from templates (some stuff you do every day), etc.

I also have an MCP server that allows you to plug it into your favorite LLM.

New comment by regnull in "Leatherman (vagabond)"

regnull — Fri, 19 Sep 2025 13:17:06 +0000

Fun fact: you can visit his cave in Ward Pound Ridge Reservation, Westchester county. Interesting place. Not many people around. You can sit in the cave and try to imagine the life he lived.