Posts on Hi, I'm Stephen 👋

How is GPT 5.6?

Sun, 12 Jul 2026 12:00:00 -0500

It’s great. However, like all evals, the verdict depends on the tasks - so I’ll describe the tasks too.

Before GPT 5.6: Fable orchestrating GPT 5.5

Right before GPT 5.6 came out, Fable was available on Claude subscription plans. I already had a $100/month Codex subscription. For Claude, as I only had a $20/month subscription, I could only use Fable in the following way:

Fable set to low
Fable using this skill to actually use GPT 5.5 as a researcher & coder

I found this paradigm on Twitter. To figure out if this was better than just using GPT 5.5, I ran two evals:

Kickbacks.AI

Thu, 11 Jun 2026 17:31:15 -0700

The receipts to show I’m not capping

First, before you jump in on this, know that they haven’t setup their Stripe integration yet. So even if you make a ton of $, this might be a scam LOL. So it’s not entirely worth it.

I ran a ton of experiments trying to go cash-positive on token usage vs. ad revenue. The idea is that they ads on Claude Code / Codex’s thinking spinner, and you get 50% revenue share. A surprising amount of companies signed up for this which is funny. A lot of them are trying to hire me.

Can AI Agents build a graphics pipeline autonomously?

Sun, 01 Mar 2026 13:51:44 -0800

I started working on adding raytracing (RTX) to The Legend of Zelda: Ocarina of Time. As in my other posts, I’m generally trying to drive the project via autonomous LLM agents.

I was hoping the project could be as simple as running Claude Opus 4.6 in a loop, saying “implement RTX into this engine, make no mistakes.” I got surprisingly far with this technique. However, I started noticing that it hit a wall, and that it seemed to be bottlenecked by its ability to understand the engine’s screenshot output, identify issues, and figure out what to work on next.

How to solve hard problems with AI

Sun, 22 Feb 2026 19:40:11 -0800

This is probably something I’ll be updating over time but has helped me a ton when kicking off long-running agents.

Find a hard problem. Something like creating a browser, adding raytracing to an old game, adding VR to an old game, decompiling an old game (are you starting to see a trend?)
Get an agent structure that allows you to run an agent on an always-on machine. It should be the same platform as the machine that will run the “problem” - if it’s a game, probably Windows. I’m using Opencode + Kimaki to check in on Discord
Ralph loops. I created a skill here. For the initial codebase, have it create a comprehensive PLAN.md, and then setup a Ralph Loop until the PLAN.md is complete. You can have it keep kicking off agents until there are no - [ ] left in the document.
(But not necessarily last): make it so that the agent can somehow verify its output. A screenshot harness, the ability to directly set state to a certain position, are super super important as as your topic gets more niche, AI is less likely to output code that works on the first try.

AI is now a magic decompiler

Tue, 10 Feb 2026 15:25:33 -0800

AI Agents are now magic decompilers. Previously, I ran an agent-in-the-loop to try to decompile Super Smash Bros. Melee in Dec 2024 with gpt-4o, but found that the model didn’t tend to learn from its mistakes. Since then:

Models have gotten smarter
Tools have gotten better
Other people have started to create skills & tools to help AI

It’s not just a couple of functions here or there. I’ve merged around 20 functions, and have 80 more in review. Previously, it would take me 1 day a function. My results aren’t even the most impressive though. The writer of the decomp skills set a record for the most matches in a PR. Tons of people are using their own custom agents or just Claude Code in the Discord (channel is #smash-bros-melee) every day with great results. If this is something you’re interested in, come pop on by! We could always use more tokens 🤠.

Re: Introducing...the Scrub

Mon, 13 Oct 2025 09:18:17 -0700

Original Article

I’ve felt this tension between “scrubs” and people trying to have fun, especially when playing competitive Smash. I do agree that it’s frustrating when people purport to be competitive but are actually scrubs. There are some points I think are missing though:

Games can change over time, especially when you complain a ton. Within the last two decades in Melee, wobbling was discovered (2006) and then finally banned (2019). The technique was degenerate, and hated in almost all levels of play. If people just sucked it up and dealt with it, I don’t think it would have been banned. If you’re not familiar, wobbling results in you being unable to move and pummeled until you die for up to minutes at a time.
Being competitive is great; but it’s not like being competitive and “having fun” are more or less honorable than each other. When you engage in an activity with someone else, the goal isn’t always to “completely crush them.” Sometimes, it’s just to have fun, right? In the case where some techniques aren’t “fun”, it seems fine to limit yourself from using them.
- If anything it’s a bit socially awkward if you can’t sense the mismatch in competitive drive and don’t tailor to what your gaming partner is expecting.

Hollow Knight: much better with a guide!

Sat, 27 Sep 2025 13:07:30 -0700

Score: 8/10

Hollow Knight: Silksong came out recently, and I got so much fomo as I had never completed the previous entry. I’ve attempted to clear the game 2 times, and just started my third playthrough. As of finishing this entry, I beat the final boss and ended up getting the 2nd best ending.

Hollow Knight is hard. If you only have a cursory interest in video games, I don’t recommend the game. As someone who is mildly masochistic and isn’t exactly good enough for Elden Ring, Hollow Knight is at the perfect threshold in difficulty for me where I can clear bosses after around 10 tries and a crashout in my journal.

`opencode` or Claude Code?

Fri, 29 Aug 2025 15:32:29 -0700

Just so that people don’t get confused, this opencode. I’m not a shill I promise.

TL;DR: if you have time to experiment, use opencode with sonnet-4. Otherwise, use Claude Code.

I’ve spent a lot of time with opencode as well as Claude Code. I’m going to use this as a live document talking about the tradeoffs of using either tool.

First, Claude Code is roughly SotA for a terminal AI editor for fullstack work (my domain). I’ve also tried:

Learning to read historical texts with empathy

Thu, 02 Jan 2025 14:55:07 -0800

Reading the Bible in an academic fashion has taught me a skill that has been extremely useful when reading other historical texts, but seems to be something that modern readers seem to be losing. I notice this especially when people struggle to read anything older than X year (e.g. 1960), because it has concepts that they fundamentally disagree with.

I call it “empathy for the author.” When reading a text and trying to understand it, my first goal is to try to learn as much about the author as possible and their worldview. Then, I attempt to understand the points they’re making. I think other people should do the same.

ChatGPT isn't a decompiler... yet

Tue, 10 Dec 2024 00:00:06 -0800

Previous article: What I’m up to

Abstract / Results

It feels a bit pretentious to open a blog post with an abstract. However, I wanted to communicate up front concisely what I tried to do, and what the open areas of exploration are. Those who are interested can dig more.

I wanted to make ChatGPT into a magic decompiler for PowerPC assembly to supercharge the Super Smash Bros. Melee (“Melee”) decompilation project. I observed over a year ago that ChatGPT was surprisingly good at understanding PowerPC assembly language and generating C code that was logically equivalent. I also saw other papers that were attempting to use LLMs as decompilers.

What I'm up to

Mon, 23 Sep 2024 17:05:52 -0700

A lot of people have been asking what I’ve been up to since I left Plaid at the beginning of this month. I was at Plaid for 4 years, which were amazing and I am very thankful for the amazing people I’ve met and work I’ve been able to do.

I am not funemployed, and I don’t want to evoke concepts related to that. I’m grinding harder than I did while employed. It’s such a gift to be able to have software engineering skills that have been forged in a real tech company, and then let loose on personal projects. I’m working on learning as much as I can about the AI space and debating if I should make that my next 4-year move. AI has been moving faster and faster, and there are so many toy projects I want to build:

Choosing a Blogging Platform: Aesthetic and Technical Considerations

Sat, 14 Sep 2024 11:57:45 -0700

In my journey as a blogger, I’ve published posts across platforms like Medium, Substack, and other proprietary blogging stacks. When writing more and more technical stuff, I realized that some stacks were definitely better than others.

When consuming other people’s blog posts, the first thing that stood out to me was aesthetics. You get an impression about the platform and the person simply by the details of how their text looks. Does their code have great, language-specific highlights? Do they use monospace + does their platform support it? How is the image formatting? What about the base color scheme?

Why pay for Notion’s AI? I built my own auto-tagging tool in a week!

Wed, 04 Sep 2024 12:55:02 -0700

Originally posted on Medium.

I built the thing I talk about in this blog post — if you want to check it out, it’s here!

Notion, like every tech company, has been shoving AI features down our throats for the last couple of months at the cost of customer UX. So I disabled them. You can do this yourself by just messaging support — I got the idea from this Reddit thread, which is one of many. Ever since I disabled it, the UX has at least returned-to-normal, and performance of editing has increased (have you ever noticed how Notion lags a bit every time you press SPC so that it can show you the AI toolbar?).

How I do window management in Mac OS X

Fri, 22 Mar 2024 12:50:14 -0700

Originally posted on Substack.

Brief doc I’m sending to my friendos

For transparency, I’m going to recommend a paid window switcher I use for Mac OS X called Contexts. It’s saved me so much time and has made using my computer a breeze; so much so, that I’ve bound it to CMD+Tab. I’ll attempt to justify this in the doc.

The default window management paradigm in Mac OS X, for me, left much to be desired. I grew up using Windows, which has a pretty different pattern for how you Alt+Tab between windows.

I made a web app to get better at adding half-steps to notes

Wed, 27 Dec 2023 12:39:25 -0700

Originally posted on Substack.

Try it out here if you like pressing buttons as much as I do! GitHub if you like reading code.

During the holidays, I wanted to get better at answering questions like “what is 7 half steps up from A?¹” I often found myself in the situation of having these problems as a lot of guitar chord sheets are written something like “A capo 7” which means you put the big barre thing on your guitar and play an A-shape chord. When using a capo, the actual underlying chord is ‘A + 7 half steps’. This means if you’re collaborating with another instrument or with someone not using a capo, you need to communicate the actual chord you’re playing. This requires some mental math, which I found slightly embarrassing as I didn’t always immediately know what chord I was playing.

The first use of YHWH in the Bible is pretty odd

Tue, 10 Oct 2023 21:18:22 -0700

Originally posted on Substack

You know the first verse of the Bible, right?

In the beginning, God created the heavens and the earth.

Gen 1:1 NET

The word for God here is Elohim. There are some interesting tidbits about this:

The noun is in its plural form. Singular would be Eloah. While this seems really controversial, in most of the usages, it’s paired with a singular verb. AKA “created” up there is masculine singular. So it’s often interpreted as a majestic plural.

Yom: Examining Creations Timeframe With Linguistics

Wed, 14 Jun 2023 21:07:37 -0700

Originally posted on Substack

I’ve often encountered Christians that believe the world was created in 7 24-hour days, because “that’s what the Bible says” in Genesis 1 and they won’t believe otherwise.

There are a ton of arguments against this “fact,” but I think the most compelling one is the linguistic ambiguity for the word “day” (yom) used in the creation account. First, let’s take a look at an English translation for one of the days of creation:

My Notes on Google's TrueTime

Sun, 04 Jun 2023 14:41:15 -0700

Originally posted on Substack

edit: this blogpost was initially wrong when I published it. Thanks to some comments I got, I learned I didn’t fully understand TrueTime or Spanner — I’ve spent some time learning and understanding the core concept again, and have updated this artifact. This is an externalized resource for me that I hope can be helpful for others.

When I was reading the famous paper on Spanner, Google’s globally distributed linearizable database, I really struggled with the concept of TrueTime, which is a core component of why they were able to get their guarantees. After trying to wrap my head around it, I created the following artifact (IMO, TrueTime deserves a mini-ish paper or post on its own):

Notifications Suck

Fri, 02 Jun 2023 14:38:05 -0700

Originally posted on Medium.

they really do. I have most of them turned off on my phone.

first, idk who to blame about these, but there is this class of notifications that are like

“feeling hungry? order UberEats now!!”

someone gave someone this lever to pull to increase engagement. the first someone upsets me, and i sometimes think about the other someone and also get angry.

i turn all of these off, when I can actually turn them off. in the case where an app doesn’t let me disable them, I turn off all notifications for that app. they’ve lost my trust.

Some quick git (and zsh) workflow optimizations

Sun, 25 Dec 2022 14:32:18 -0700

Originally posted on Medium.

I use git a ton in work and my personal life, and have come up with a couple of aliases that have made using it so much more pleasant and fast. First, I’ve renamed git to g. You have no idea how much typing that’s saved me (I also don’t know).

alias g="git"

Aliases

git allows you to alias commands. In the spirit of renaming commands to one character, here are my favorite aliases:

I’m against AIs writing blog posts

Sat, 03 Dec 2022 14:28:29 -0700

Originally posted on Medium.

Why do we need AI to write blog posts? There’s been a trend of these AI tools to write things for you:

Imagine you have something to say to people. So you come up with some representation of that thing you want to say; maybe bullet points, maybe a rant, maybe even some markdown stuck in a GitHub repository. Why can’t you just publish that?

Frontend developers: stop moving things that I’m about to click on

Fri, 25 Nov 2022 13:58:48 -0700

Originally posted on Medium.

Please. It’s an insult to my brain. Like, you put the button in one place, and then you’re like “nah, let’s move it somewhere else.” Here are some examples that have annoyed me the most lately:

Lyft Bike Scan Button

This one sucks as I’m often opening the Lyft app just to ride bikes. So I immediately hit the bike button and I’m trying to hit the “Scan” button as quickly as possible. But no! Depending on how fast my internet is at the current moment, that banner will appear and push the Scan button up, which means I end up hitting the banner instead 😟; this is often a really frustrating start to my daily commute.

Big Rocks

Wed, 23 Nov 2022 13:56:54 -0700

Originally posted on LinkedIn as a meme.

i’ve been thinking about the big rocks every day

for those who don’t know: https://lnkd.in/gZGfgBE9

just the other day, i was at whole foods. and i tried to fit the items into my cute tote bag haphazardly. they didn’t fit!!

but then, i put the straus whole fat milk first. and then the yogurt. and then i put the cheese, and the cute lil snax. i fit everything! the bag was kinda heavy though.

Attachment Styles at Work

Sat, 17 Sep 2022 13:50:20 -0700

Originally posted on LinkedIn as a meme.

my friends are all attachment styles this and attachment styles that when it comes to dating. however, has anyone had the brilliant idea of applying attachment styles to working at a tech company? I didn’t think so! now you might be thinking, “hey Stephen isn’t this a bad idea?” and yes, it is. but just like we should all hope to be secure partners in our relationships, we should also possibly strive to be secure capitalistic partners at work.

Adopting the OpenAPI schema to generate Plaid's SDKs

Wed, 15 Sep 2021 13:03:41 -0700

Originally posted on the Plaid Engineering Blog while I was part of Plaid’s Developer Experience Team.

The developer experience team focuses on building tools and features that make it as easy as possible for developers to explore our APIs and integrate with Plaid. This year, our team adopted an OpenAPI schema (OAS) as a specification for our API. We launched this schema in beta earlier this year.

Our team is responsible for maintaining three sources of truth for our API that allow developers to build and test their integrations:

Writing ES6 Everywhere

Sat, 28 Sep 2019 12:55:44 -0700

Originally posted on the Mobile Developers of Berkeley blog

This article is going to be a quick intro to the basics of writing modern JavaScript all the time, rather than being dependent on what environment is supported by a given browser.

Intro

First, what is ES6? ECMAScript 6 is the sixth standardized version of JavaScript, which is ultimately a specification of language features. ES6 added a ton of really great features that drastically improved the ability to construct larger-scale programs with JavaScript (like constants and block scoping!). It’s important to note, as ES6 is only a specification, it’s ultimately up to the browsers to provide an implementation for these new features: often, certain browsers lag behind on implementing all these features (i.e. IE).

A Quick Dive Into Python's Slots

Fri, 09 Nov 2018 18:00:10 -0700

Originally posted on Medium.

For my first Medium article, we’re going to go into a quick and easy way to speed up your Python code (and pass those pesky HackerRank tests where you’re just a bit short on time!), as well as some of the technical implementation details for the curious.

__slots__ is an attribute you can add to a Python class when defining it. You define slots with the possible attributes that an instance of an object can possess. Here’s how you use __slots__:

Posts on Hi, I'm Stephen 👋

How is GPT 5.6?

Before GPT 5.6: Fable orchestrating GPT 5.5#

Kickbacks.AI

Can AI Agents build a graphics pipeline autonomously?

How to solve hard problems with AI

AI is now a magic decompiler

Re: Introducing...the Scrub

Hollow Knight: much better with a guide!

`opencode` or Claude Code?

Learning to read historical texts with empathy

ChatGPT isn't a decompiler... yet

Abstract / Results#

What I'm up to

Choosing a Blogging Platform: Aesthetic and Technical Considerations

Why pay for Notion’s AI? I built my own auto-tagging tool in a week!

How I do window management in Mac OS X

I made a web app to get better at adding half-steps to notes

The first use of YHWH in the Bible is pretty odd

Yom: Examining Creations Timeframe With Linguistics

My Notes on Google's TrueTime

Notifications Suck

Some quick git (and zsh) workflow optimizations

Aliases#

I’m against AIs writing blog posts

Frontend developers: stop moving things that I’m about to click on

Lyft Bike Scan Button#

Big Rocks

Attachment Styles at Work

Adopting the OpenAPI schema to generate Plaid's SDKs

Writing ES6 Everywhere

Intro#

A Quick Dive Into Python's Slots

Before GPT 5.6: Fable orchestrating GPT 5.5

Abstract / Results

Aliases

Lyft Bike Scan Button

Intro