HN user
Certified Public Accountant turned fintech engineer. I used to build financial tools & technology for a credit card company whose product you may have in your wallet. Now, I build credit infrastructure at Nova Credit.
Court rules employee has the right to make the comments she made
Random HN comment:
her comments were unprofessional and she has no right to distract the company.
Same User:
y Downvote :(
I’m building a blackjack card counting tool for people to learn how to count and how to identify games that are winnable. It is designed to take a completely novice to an advanced, winning card counter, using a Duolingo like approach - mastery based learning across sequential modules. Minus the ads and dark patterns.
This is the proper approach when deciding whether to use any type of tool or technology. Is the increased amount of cognitive overhead for someone with minimal exposure to your system (who will have to maintain it when you’ve moved on) worth the increased performance on a dollars-per-hour basis? If so, it may be a good option. If not, it doesn’t matter how much better the relative performance is.
Bank Rank — an automated bank account that dynamically allocates your money across the best accounts to maximize your rate at all times.
We’re doing an alpha launch in Q1 2026, and if you’re interested, sign up at bankrank.io/waitlist or email bankrank.alpha@gmail.com
Two big reasons:
1. If the government is in charge of deciding the tax policy and collecting the taxes, it creates a potential conflict of interest if they are also in charge of telling you how much you owe. In theory, they could charge you more than they're legally allowed to, but how would you know unless you (or someone else) also calculated your taxes? A common suggestion to this is to have the government give a return that shows what they _think_ is owed, but this creates a conflict if the government accidentally underbills you, since you're not likely to correct the mistake. In order to ensure compliance on both sides, both the government and individual need to prepare the tax return. Otherwise, one party risks being overcharged/underpaid.
2. Tax evasion is an effective law enforcement tool for catching criminals, so by putting the burden on the individual to report taxes, you add another tool in the law enforcement toolkit. From the state's perspective, it is more compelling to tell a jury "this person owed $5 but only paid $1" than "this person owed $5, but only paid $1 because we told them they only owed $1." Tax evasion is how famous gangsters like Al Capone and other shady-characters have historically been caught[0]
The tax prep industry is lucrative largely because of lobbying and consumer ignorance. There are plenty of free-file options for folks below certain income thresholds, as well as non-profits who will do your taxes for free. There are also lots of free tax-prep sites, but they are being drowned out by the advertising and lobbying of the for-profit tax-prep industry.
To add my own 2-cents: if your income comes from investments, 1099, or W2, you likely can do your own taxes in about an hour. I personally use TaxHawk [1] since it's free for federal and $16 per state return, and has the same kind of interface as turbotax and the like. If you want to save on that $16, you could use TaxSlayer [2] instead -- I've used all of them, and personally prefer TaxHawk. Just remember to decline any of the upselling they do just before you submit your refund. You probably don't need the premium service, a dedicated tax pro, nor audit protection.
Source: am a CPA
[0] https://en.wikipedia.org/wiki/Al_Capone#Tax_evasion
I’m building a tool to automatically move deposits into new bank accounts with better rates and fees. It will operate continuously, so I never have to search for the “best” bank account — I’ll always have it. Think wealthfront for checking and savings.
Are you able to share a bit more about this? Specifically:
1. How did you find your first customers?
2. How did you determine a price?
3. What are the payment plans based on? Usage? Flat rate? Something else?
4. How much are you currently earning MRR and what are your costs like?
I work in software now, but I was an accounting major, CPA, and worked in Public Accounting for 4 years before making the switch.
Accounting actually is more of a trade, and I felt like the majority of classes they taught helped me on-the-job as an accountant. People who like systems engineering would enjoy accounting, because it's getting to the nuts-and-bolts of our financial system and understanding both the how and the why. Whether you're international conglomerate Apple, Inc or software engineer calderarrow working a day job, the laws of accounting still apply.
But I would not recommend paying more than you need to for an Accounting Degree. The Big 4 will recruit anyone from anywhere, and as long as you have a degree (which is usually required for the CPA License) and a 3.3+ GPA, you can get a job in any major city. Assets + Liabilities = Equity whether you're at community college or Harvard, so get the cheapest degree you can get.
Also, try to squeeze 150 credits into your undergrad curiculum, as some CPA licenses require 150 credit hours. This is typically 4 years of undergrad (120) + 1 year of a master's (30). The added cost of tuition for those classes isn't that much, but being able to get out and start working a year sooner is a ~$60k decision that is worth it if you can do it.
As for why I left accounting: I started learning Python to help me automate some of the boring parts of my job [0] and fell in love with software. It just clicked for me in the same way that accounting did, and now I work in fintech, where I'm able to blend the two.
With capital punishment, we remove the possibility of fixing our mistake if someone is innocent. But if we lock someone up for 30 years, they have the opportunity for justice to be served, and we have mechanics to _try_ to make it right.
Perhaps I misunderstood it, but I read that as you not finding the value of avoiding accidental executions as worth the cost to avoid them. And if so, then I think that’s a perfectly valid position. But it’s subjective, since others may find the value worth it.
I’m curious though: is there an error rate where you would feel like capital punishment would be off the table? For instance, if 90% of people executed were innocent, would you still want it for the 10% who deserve it? I admit that if we had a 100% success rate, I would be open to capital punishment, so we may actually agree that there’s a threshold where the system shouldn’t be allowed to use that as a form of punishment, and only disagree about about the percentage.
I used to be, for most of the same reasons as you. What ultimately convinced me was realizing that our judicial system can never be 100% perfect, so we would always have a non-zero number of innocent people executed as long as capital punishment is on the table. To me, I think the cost of keeping people incarcerated is worth the cost of accidentally executing innocent citizens.
Put a bit more personally: would you support capital punishment if you had to pull the trigger, and you would be killed if you executed an innocent inmate? Most people I speak with would be fine pulling the trigger, but no one I’ve talked with would be okay with taking responsibility for mistakes.
Off-topic, but did anyone attend Johns Hopkins CTY camp as a kid? I did it for 9 straight summers, and while the name is a bit pretentious, it was one of the best experiences I had. I am curious if anyone is an alum, since the overlap of CTY and HN seems pretty unique atm.
That section is just for mathematics, which the commenter specifically called out. The reading and science are above average
If you don't mind sharing even more, what did you do to learn HPC/AI/ML? Any suggestions for getting started?
Give that money to someone close to you who can hold you accountable — hell, I’ll do it if you need — but you’re going to be earning every dollar of that over the next year by counting calories. Here’s how:
It’s approximately 5000 calories per pound of weight that you want to lose. 30 pounds * 5000 calories = 150k calories to lose 30 pounds. That’s pretty close to $100k Pay yourself to count calories, $1 per calorie. Track everything. Every meal, every snack, everything you put into your body. Every week, add up the total calories consumed and subtract the total amount from your weekly “maintainence” calories. To calculate that, multiply your current body weight by 16 or 18, depending whether you exercised that day. For instance, if you weigh 200 pounds, you should eat about 3200-3600 calories daily to maintain that weight, closer to 3200 if you don’t exercise, 3600 if you do. Every week, pay yourself $1 for every calorie you burned, re-weigh yourself, and repeat.
No exercise needed, but that will definitely accelerate the process, but only so much. It took months to get you here, so expect to spend months to get you out.
SEEKING VOLUNTEERS:
Project: Bank Rank
Description: Crowdsourcing templates to monitor deposit account rates and fees across the 8500+ banks and credit unions in the U.S.
Tech Stack: Vanilla Javascript (Playwright/Puppeteer)
Github: https://github.com/project-bankrank/bankrank
I was going to make a separate post about this in a few days once I finished up the documentation, but I'll post it early even though the Github repo is not in a completed state. The general idea is to build a tool that can scrape bank account information whenever it is executed, to enable more transparency when shopping for financial products like savings accounts. I've started working on the engine and templates, and currently have the logic built out for several hundred products at ~70 financial institutions.
The best way you can help is by creating "templates" of various product information by recording or manually writing puppeteer/playwright scripts to extract specific information about various bank products. This is a simple task, but due to each bank having multiple product offerings (checking, savings, money market, certificate of deposit) and multiple tiers of each product, the actual market for a "savings accounts" is in the tens of thousands. Banks change rates and fees a lot more frequently than they change their websites, so even though this is a brute-force method of aggregating data, it will be sufficient for most consumers.
If you're interested in learning more, shoot me an email: project.bankrank@gmail.com
I've been trying to understand the rationale of the court in simpler terms. Is this an adequate understanding of the reasoning:
1. Congress delegates some of its authority to issue/handle student loans to the executive branch via the US dept of education and some legislation passed in the past.
2. As President and leader of the executive branch, Biden wants to utilize the authority granted to him to modify the terms of the loans due to the impact of the COVID 19 national emergency. [0].
3. As part of his loan modifications, he wants to forgive a certain portion of the loans altogether, for which he was sued.
4. The Supreme Court ruled that the modifications of the loan forgiveness were ultimately unconstitutional due to the major questions doctrine, implying that while Congress may have delegated some authority to the executive branch for managing loans, outright forgiveness on such a scale would be considered economically significant, and therefore would be presumed not to be delegated. [1].
Am I missing something?
If people take their money elsewhere, the banks can borrow from the fed, just at a higher rate. So in theory, they should be willing to increase their rate up to whatever the fed-rate is, minus some margin for the risk of dealing with lower sums of deposits.
There's a direct correlation between advertising spend and interest yields on deposit accounts, which is somewhat interesting. It seems like the optimal business model is to focus more on customer acquisition than customer retention, because these products are very sticky. Churn rate for deposit accounts is like 8%, which is crazy given the rate variance.
This is a step in the right direction, and personally I think that low-level deposit accounts are ripe for serious competition. There are thousands of banks and credit unions (~9,000 if memory serves) all offering the same risk-free product, but the rates and fees can be wildly different.
The FDIC publishes some decent statistics about average yields for various deposit products.[0] As of March 2023, the national average for savings accounts was 0.37%. The largest banks (Wells, BoA, Chase, Citi) offer 0.01% interest. That's 1 penny for every $100 you loan them.
That low rate made sense when the federal funds rate was low as an incentive for banks to borrow from consumers rather than the government. But now the rate is almost 20x more than it was a year ago, so there's a larger incentive for banks to borrow from consumers.[1] This should drive more competitive rates offered for consumer deposits.
If you're earning less than 1% APY on your checking and savings account and have $1000 in it, spend an hour researching an alternative. Going from 0.01% to 1% alone would earn you $10 more per year on that amount. If you have $10k, that's an extra $100 for maybe an hour of your time.
[0] https://www.fdic.gov/resources/bankers/national-rates
[1] https://fred.stlouisfed.org/series/fedfunds
---
On a related note, I'm part of a team working on a service to monitor all the deposit rates across all financial institutions in the US in real time. If you're interested in learning more or signing up for the alpha launch (targeting Summer 2023), shoot us an email at bankrank.alpha@gmail.com
Fellow software devs & CPAs unite! Thanks for the kind words.
There aren't many of us, so I’d love to connect. If you want, shoot me note at: Anthonyj at gwu.edu
I’m very excited to see how technology impacts financial reporting in the future. We’re rapidly approaching a point where every single transaction could be audited in real time with software, and the details of each transaction automatically scrutinized.
Auditors have access to all the financials, but they only audit a statistically significant sample, because it would be incredibly expensive to audit every transaction.
Fraud can be easily detected if one employee is committing it. Fraud is substantially harder to find if two employees are involved, specifically 2 employees involved in internal controls.
For instance, if you have a policy that all checks paid over $10k require 2 signatures from corporate officers, it’s easy to catch a check with one officer forging the name of a second in order to siphon money to his 3rd party shell company.
But if both officers make a shell company, they can post the check as usual, and the check would pass auditor checks unless they looked into the specific corporation being paid, which may be out of scope if it’s a relatively small transaction.
Ultimately, you don’t need assurance that the financials don’t have fraud, you want assurance that they’re materially correct. Whether the company lost 10k to fraud or waste or incompetence is almost irrelevant for the investor, because the company has 10k less money. Obviously they’d prefer it not be due to fraud, but the impact on the financials is more or less the same.
Source: am a CPA
The links referenced in that post are broken, but I would be interested in seeing the data on this.
As a personal anecdote: I remember being at the ESRI Developer Conference back in 2019, and one of the keynote speakers spent a good amount of time talking about whether we were software "developers" or "engineers." I'm paraphrasing a bit, but ultimately he concluded that we were engineers because we follow scientific methods, identify problems, solve them quantitatively, and create solutions within the confines of external constraints, such as the laws of physics.
During the Q & A, someone stood up and proposed that by his definition, authors were literary engineers, since they approach writing a novel in the same way that we approached writing code.
I didn't have a horse in this race, but I remember thinking about how silly it was for my company to spend hundreds of dollars to fly me out across the country so I could listen to brilliant people discuss semantics at a software conference.
If you'd like, I'll send you an update when we go live. Feel free to shoot me an email: bankrank.alpha@gmail.com
That could work! Are there any specific integrations you'd want to see which we could prioritize?
Thanks for the feedback! To be fair, we're definitely not capturing a sizeable portion of the banks/credit unions that are actually "good". We prioritized the larger banks, but we definitely intend to capture every bank/cu down the road.
As far as the results go, we figured that users would want to see how they can break up their deposits across multiple accounts to maximize the yield of all their deposits. So putting 1k in DCU may be pretty nice for someone with 3k, since that 6% is a massive return on what's effectively a risk-free product. For someone with more savings, I could see how that drop in the bucket is pretty insignificant, and could probably be filtered out.
A common theme of the early feedback we've gotten has been "add more filtering," so it seems like it'd be good to have a way to limit the results to only the best single account that could take the entire balance. We'll work on that.
Also, thanks for the heads up about HMB and OF. We'll add those later this week!
Thank you! That’s a particularly good rate, so we’ll prioritize those this week
I actually never heard of them until we started working on this project. I was surprised to see their rates of like 6%.
It’s only on the first $1000 or something, so it seems like a promotion to get them listed on those “top 20 bank” listicles. Regardless, I plan on opening a savings account with them because that’s ridiculously good for a risk free product