Hacker News Reader

pyrrhotechon Ask HN: What apps have you created for your own use?

An algorithmic trading platform. I ran it successfully for a couple years with my own capital alone before opening it up to others (https://grizzlybulls.com). It's been quite successful compared to the S&P 500 since launching publicly in Jan 2022. Performance of our various models vs. the benchmark below:

SPX (benchmark): -3.31%

VIX TA Macro MP Extreme: +29.91%

VIX TA Macro Advanced: +23.08%

VIX TA Advanced: +3.42%

VIX Advanced: -0.67%

Vix: -4.18%

TA MR: +10.12%

TA T: +2.3%

pyrrhotechon Almost no one pays a 6% real-estate commission except Americans

I'm an American that has bought and sold 3 houses and never once paid the 6% commission. It seems lots of folks don't realize commissions are negotiable, and that discount brokers exist who charge much less for roughly the same end product. If you think the standard full service 6% is providing you adequate value, by all means pay it, but don't feel like you have to.

pyrrhotechon Ask HN: Ways to make side income as a SWE?

yea if you want to go it the investing route, you need much better returns than 4% real. Personally I've been beating the market with algorithmic trading (https://grizzlybulls.com/models/vix-ta-macro-mp-extreme) but it's still a long game unless you start with millions

pyrrhotechon Ask HN: Ways to make side income as a SWE?

yes but making 7 figures annually is quite a bit different than simply having a 7 figure total net worth. The latter is achievable by most frugal middle class Americans over the course of a career, often much shorter. The former is usually not possible through a day job over a normal career length

pyrrhotechon Ask HN: Ways to make side income as a SWE?

what if you desire 7 figure income? very few full time jobs in this field would cut it. Only options are to either give up dreams and accept mediocrity, start a successful side business, or possibly work 100 hour weeks stacking multiple full time jobs

pyrrhotechon Hard-to-swallow truths they won't tell you about software engineer job

I would say upper class is about net worth, not income, and probably top 1% or so would qualify

pyrrhotechon Hard-to-swallow truths they won't tell you about software engineer job

Here's another good one. You almost certainly will never make it to the upper class with this career, no matter how well you perform, unless you transition to management and climb to VP+ in a large company or start your own company. You can be comfortable as an engineer, but it's not the gold mine lots of folks think it is.

pyrrhotechon Nvidia announces financial results for second quarter fiscal 2024

He's worth way more than that. https://www.bloomberg.com/billionaires/profiles/jenhsun-huan...

pyrrhotechon Grindr employees have 2 weeks: agree to move across the country+RTO or lose jobs

Imagine having the economic performance of France vs. the United States

https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?location...

pyrrhotechon Is technical analysis just stock market astrology?

TA is unlikely to produce alpha when used on its own, but it's certainly possible. However, combining TA with other indicators such as macroeconomic data, VIX futures curve, and tracking interest rates and Fed actions can lead to a consistently profitable model, as long as overfitting is kept in check and slippage is properly accounted for. It's much easier said than done.

It took us years of 80 hour weeks to become consistently profitable at Grizzly Bulls, but the result is quite satisfying. Our top model, https://grizzlybulls.com/models/vix-ta-macro-mp-extreme, has produced +17.92% returns since launch in January 2022, vs. -4.15% for the S&P 500 over that same time period. TA plays a role in timing swing trades / hedges once other indicators align.

pyrrhotechon Case study: Algorithmic trading with Go

I've also just started a blog series for getting started with algorithmic trading in NodeJS for complete beginners: https://grizzlybulls.com/blog/building-an-algorithmic-tradin.... A lot of folks would raise an eyebrow at the thought of using NodeJS for algotrading, but I've found it to be a great fit for my non-HFT, more swing-trading style, and I've used it successfully in production for over 3 years.

pyrrhotechon Case study: Algorithmic trading with Go

The models vary greatly in the number of external data sources they pull from to compute signals. The simplest ones like https://grizzlybulls.com/models/ta-mr-basic and https://grizzlybulls.com/models/ta-trend-basic only use a technical analysis calculations on the raw price and volume data of SPX on a few different timeframes. The most complex pulls data from dozens of external sources as it incorporates macroeconomics, monetary policy, various market sentiment sources, yield curve, valuation data, fundamentals, earnings estimates, etc.

It's also worth noting that every additional data source adds some risk of that data source being down or publishing inaccurate data during real-time signal calculation which can cause inaccurate signals, so in order to justify that risk, the external source must meaningfully contribute to alpha or better risk-adjusted returns.

None of them involve a human element in real-time. However, they are occasionally updated as new data comes in, but any updates only apply going forward so as to preserve the live trading history accurately (live trading start date varies by model from mid 2020 to jan 2022 with 2009 - 2020 being purely backtest for all models).

pyrrhotechon Case study: Algorithmic trading with Go

Thanks for the feedback. I try to call out in several places in the app that 2009 - 2020 is backtest only, but perhaps I need to make it more clear. No one should ever expect a model to trade as well in live trading as it backtests, and that's been true of nearly all our models (save the anomaly of TA-MR-Basic). However, there's more than enough room in the returns and drawdowns to underperform the backtest while still producing significant alpha, which is what we've seen with 4/7 models and especially with the top two.

However, the April 2009 start date is not actually random--it's the first start date for which intraday futures data is available for more than just front month contract. Several derivative indicators of the VIX futures curve are the most foundational to all the VIX-based models, and they simply cannot be processed without it. The VIX futures were only created in 2004, and I've scoured the internet for intraday data for more than just front month (can't create the curve if you only have front month data), and the earliest it can be found is April 2009.

pyrrhotechon Case study: Algorithmic trading with Go

One of the most misunderstood concepts about algorithmic trading is that speed is not a critical factor for the majority of systems. My systems like https://grizzlybulls.com/models/vix-ta-macro-mp-extreme have been crushing the market with live trading for 3+ years and yet trade on a frequency of only once ever 18 trading days on average (some a little more often, some even less), and are set up to only generate signals around hourly breakpoints.

The last 18 months have been weaker than the first given the enormous structural shift in the market in this high inflation and rapidly rising interest rates environment, but we've still managed to deliver a return of +14.11% since the site launched in Jan 2022 compared to -7.83% for the SPX. We've managed to do it without any use of leverage and also with lower drawdowns as well of -16.48% vs. -27.57% for the SPX over that time frame.

pyrrhotechon Bootstrapping Atlassian: $10k In Credit Card Debt To $20M In Revenue In 5 years

It's a great story, but at least in the software world, this sort of thing is exponentially more difficult to pull off today than it was 10 or 15 years ago. The internet was the most economically revolutionary invention of all time. Sell to anyone! Connect anyone in the world instantly! It created a lot of low hanging fruit. So much that it took 20 years or so to grab it all. However, in 2023, all that low hanging fruit is gone, and the market is largely saturated. A quick Google search of nearly any idea will pop up at least a handful (sometimes 30+) of established competitors fighting over scraps, none of which ever have a chance of achieving Atlassian-level success. Even B2B SAAS is reaching that saturation point. The golden age of the garage tech startup becoming a billion dollar business in a few years are probably behind us.

pyrrhotechon πfs – A data-free filesystem

the set of reals is larger than the set of integers despite both being infinitely large

pyrrhotechon Nvidia is now a $1T company

It may continue up to 1.25T or so, but the current price is absurdly overvalued at 37x sales. Even with the stellar guidance they gave, at best, the company is worth 0.5T. It may not go down that much, but folks buying here for the long term are in for disappointment.

pyrrhotechon Meta fined $1.3B over data transfers to U.S.

HN may be slightly better than other platforms, but it's still largely an echo chamber

pyrrhotechon Meta fined $1.3B over data transfers to U.S.

It's amazing to me how many otherwise intelligent people on HN inevitably make this same comment, when in fact, this is a substantial fine even to a company the size of Meta. Much higher would be borderline extortion, and Meta would seriously start to consider whether doing business in the EU is worth it.

pyrrhotechon A morale crisis after multiple Meta layoffs

Perhaps it's done great on a YTD basis or since the panic low last Fall, but overall Meta has been a laggard for many years now. A mere 36% return over the last 5 years compared to 66% for SPY and 96% for QQQ. Furthermore, it's still nearly -40% off its peak in September 2021.

pyrrhotechon Five years of indie hacking

It doesn't take extreme privilege to quit your job and indie hack for 5 years. It takes $60k in savings and extreme frugality (or the equivalent of the cheapest amount you can live on for 5 years in your country). $60k of savings can easily be saved in a year with a junior dev job in the USA, or 2 years in Europe as long as you have the frugality. Of course, you do sacrifice having a family, living in an expensive area, going out to restaurants and all other luxuries, but you can own your time without being born into any sort of wealth or privilege outside of being born in a country with GDP per capita above $10k.

Most people would rather live a more normal middle class life and work for the man instead. To each his own, but don't act like it's not a choice available to practically 100% of HN readers.

pyrrhotechon Ask HN: Are Paul Graham's Classic Startup Essays Outdated?

For the most part, all the low hanging fruit that the great startups of the last 30 years monopolized on has been picked. The next 30 years will look nothing like the last. I suspect we'll be seeing much fewer sub 50 year old self-made billionaires, a near total collapse in the number of inflation-adjusted unicorns under age 10, and a relative shrinking of what we think of currently as "tech" in proportion of total GDP. We may even soon to stop thinking of software companies as tech companies by default.

Even looking at the last 10 years vs. the 10 years before that, the emphasis has already started to shift back the power to the celebrity content creator over the nerdy engineering tech creator. Before Web 2.0 really took off, websites and apps were all the rage because it was mostly nerds and other early adopters that could monopolize the power of the internet. Now the playing field has been leveled again, and it's those with strong personalities, marketing skills and other digital content production skills who have been making the big bucks. Mr. Beast's youtube channel was offered a $1 billion buyout. Engineers weren't thought of very highly before the personal computer revolution, and the money and power was all with the celebrities instead. I think the last 30 years was mostly an anomaly, but now that the internet is more democratized, the money and power will be returning to the celebrity content creator.

pyrrhotechon My fifth year as a bootstrapped founder

Congrats on your amazing success! 0 to 850k revenue with a hardware product in three years is almost unbelievably amazing. I'm totally ignorant about hardware dev. How does one even go about designing and manufacturing a hardware product like TinyPilot? Do you design in some sort of CAD software and then partner with some sort of factory?

pyrrhotechon Ask HN: Laid off folks, are you getting hired?

Or they are looking at those juicy severance checks and deciding to take a 6 month vacation before applying anywhere again, using state of the economy as an excuse.

pyrrhotechon Ask HN: Recruiters have stopped contacting me? What could have happened?

https://www.investopedia.com/terms/d/deadcatbounce.asp

"if you throw it out the top floor window, even a dead cat would bounce". I shouldn't use the phrase though, as I'm actually very fond of cats. I think it's an example of society's bias against cats. If it were called "dead dog bounce" it would likely be taboo

pyrrhotechon Ask HN: Recruiters have stopped contacting me? What could have happened?

The tech bubble is bursting, did you miss the memo? Even after the recent dead cat bounce, FAANG stocks are down 30%+ with their revenue growth collapsing and they are laying off workers, and it's a whole lot worse for all the non-profitable zombie companies that have been surviving solely on ZIRP and Quantitative Easing absurdly loose financial conditions that have come to an abrupt halt.

Naive investors in the tech bubble forgot the hotdog stand problem: https://seekingalpha.com/article/4558508-what-hotdog-vendors... so many got burned and they aren't coming back any time soon.

pyrrhotechon Bay 12 Games has made $7M from the Steam release of Dwarf Fortress

Very well deserved, and something they should have done a long time ago. Better late than never

pyrrhotechon Why EVs Aren't a Climate Change Panacea

EVs seem to not be a very well thought-out long term solution in general, unless radically different batteries with very high energy density can soon be developed. Earth has 88 million tons of lithium, of which only 1/4 is economically feasible to mine. Each Tesla battery uses about 50 kg of Li, which means only about 18 Tesla batteries can be produced per ton, or about 396 million total Tesla batteries. Current global automobile production is about 80 million per year. If society were to totally switch to EVs, we would run out within a decade or so even assuming most of these cars used far less lithium (and thus had much shorter range, limiting practicality). This back of the napkin analysis also ignores all the other competing uses of Li...

pyrrhotechon Ask HN: What have you created that deserves a second chance on HN?

I would start with learning the basics about the primary forces that move markets (examples are earnings expectations, macroeconomic health indicators, and sentiment measurement via VIX and TA). The hard part is identifying legitimate leading indicators / patterns of combinations of these that have lasting and tradeable impact, i.e. building your trading strategy with as little overfitting of noise as possible. Actually automating that strategy is the relative easy part, though still a pain the ass because brokerage APIs are garbage and often you'll find no API exists for a particular set of data you want so you have to settle for brittle scraping. I would also check out videos from a couple respectable algotraders on Youtube such as Kevin Davey and Jacob Amaral.

pyrrhotechon Ask HN: What have you created that deserves a second chance on HN?

Mostly access to the live signals of the model unlocked by that premium tier, with varying levels of notifications to real time signal changes (email, text, API and full automation option)