Do the simplest thing that could possibly work
Read it on seangoedecke.com When designing software systems, do the simplest thing that could possibly work. It’s surprising how far you can take this piece...
Finding the low-hanging fruit in tech companies
Read it on seangoedecke.com Suppose your job is to pick fruit in a giant orchard. The orchard covers several hills and valleys, and is big enough that you’d...
Everything I know about good API design
Read it on seangoedecke.com Most of what modern software engineers do1 involves APIs: public interfaces for communicating with a program, like this one from...
Don't feed me AI slop
Read it on seangoedecke.com In the early days of any new technology, the relevant social norms are still being workshopped. For mobile phones, that meant...
The famous "bottomless pit" AI greentext is fake
Read it on seangoedecke.com Many people believe1 this is the best piece of art or humour that AI has ever produced: This was generated three years ago by...
Is chain-of-thought AI reasoning a mirage?
Read it on seangoedecke.com Reading research papers and articles about chain-of-thought reasoning1 makes me frustrated. There are many interesting questions...
What's the strongest model you can train on a laptop in five minutes?
Read it on seangoedecke.com What’s the strongest model I can train on my MacBook Pro in five minutes? I’ll give the answer upfront: the best 5-minute model I...
OpenAI's new open-source model is basically Phi-5
Read it on seangoedecke.com OpenAI just released its first ever open-source1 large language models, called gpt-oss-120b and gpt-oss-20b. You can talk to them...
AI interpretability has the same problems as philosophy of mind
Read it on seangoedecke.com If you want to know how a particular AI language model works, the current state-of-the-art approach is to use a “probe”. Pick one...
Why I don't allow AI-generated content on my blog
Read it on seangoedecke.com I will never include any AI-generated content on this website. In general I’m a big AI believer, and I use large language models...
Trying to impress engineers, trying to impress managers
Read it on seangoedecke.com In the first few years of my career, I knew next to nothing about how to do good work in a tech company. The senior and staff...
The refusal problem in AI language models
Read it on seangoedecke.com Everyone’s interested in what language models can do. But what about what they will do? Try giving ChatGPT 4o a menial task that...
Can small AI models think as well as large ones?
Read it on seangoedecke.com An AI trend that’s emerged in the last few months1 is the idea of a “cognitive core”. Instead of trying to build the largest,...
What kind of work I want (in 2025)
Read it on seangoedecke.com In 2021, I wrote this post describing the kind of work I wanted then. Almost five years later, it’s time for an update. What kind...
Do not yell at the language model
Read it on seangoedecke.com A few days ago, the news went viral that Replit’s AI coding agent went rogue and deleted a production database. A lot has been...
The silent AI war inside tech companies
Read it on seangoedecke.com I recently wrote an article for LeadDev about how software engineers are dealing with the AI hype in practice. Here’s my...
The case for regulating AI companions
Read it on seangoedecke.com In April, OpenAI screwed up by releasing a version of GPT-4o that was overly sycophantic. If you told it that Brian Cox was...
How to actually use Model Context Protocol
Read it on seangoedecke.com Everyone’s very excited about Model Context Protocol, or MCP for short. In a sentence, MCP is a universal protocol for exposing...
Practical notes on getting LLMs to generate new ideas
Read it on seangoedecke.com Large language models struggle to generate new ideas. To AI skeptics, this seems trivially true, since they believe LLMs can only...
Building a RTS where you issue orders to AI agents
Read it on seangoedecke.com I’ve always been interested by the third-act turn in Ender’s Game where the child strategists switch from directly controlling...
The three great virtues of an AI-assisted programmer
Read it on seangoedecke.com In 1991, Larry Wall famously wrote that the three great virtues of a programmer were “laziness, impatience, and hubris”. A...
METR's AI productivity study is really good
Read it on seangoedecke.com I complain a lot on this blog about AI studies. Given that, I ought to give credit when an AI study comes out that I think is...
Why we should anthropomorphize LLMs
Read it on seangoedecke.com I recently read a very good blog post by Halvar Flake titled “A non-anthropomorphized view of LLMs”. Here’s a very brief summary:...
Mecha-Hitler, Grok, and why it's so hard to give LLMs the right personality
Read it on seangoedecke.com Recently, xAI’s Grok model made some very strange comments. In a now-deleted post, it suggested Adolf Hitler as the right person...
The more senior engineers get, the more results matter
Read it on seangoedecke.com In my experience, people tend to overrate how much moving up the org chart changes the fundamentals of the job. As a staff...
How I use LLMs to learn new subjects
Read it on seangoedecke.com If you want to learn about a new subject in 2025, one of the best ways to do it is to ask a strong language model. The reason...
Building tiny AI tools for developer productivity
Read it on seangoedecke.com The two current paradigms of AI app development are agentic coding (using an AI to help you write your ordinary line-of-business...
How to become passionate about delivering shareholder value
Read it on seangoedecke.com I am passionate about delivering shareholder value. It feels kind of embarrassing to admit, but it’s true1. I like all the things...
What happens when engineers work more than one job
Read it on seangoedecke.com There’s a news story going around about an engineer called Soham1 who was routinely “overemployed”: he maintained remote jobs at...
Continuous AI in software engineering
Read it on seangoedecke.com When I use AI in my software engineering job, I use it “on tap”: when I have a problem that I’d like to run past the LLM, I go...
I don't care if my manager writes code
Read it on seangoedecke.com I write a lot about how the post-2010s economic squeeze has made it harder to be a software engineer at a large tech company. But...
AI coding agents are already commoditized
Read it on seangoedecke.com All of a sudden, it’s the year of AI coding agents. Claude released Claude Code, OpenAI released their Codex agent, GitHub...
Pure and impure software engineering
Why do solo game developers tend to get into fights with big tech engineers? Read it on seangoedecke.com Why do solo game developers tend to get into fights...
Everything I know about good system design
Read it on seangoedecke.com I see a lot of bad system design advice. One classic is the LinkedIn-optimized “bet you never heard of queues” style of post,...
What's going to happen to junior engineers?
I'm worried about the impact of AI on the market for juniors. Read it on seangoedecke.com In the 2010s, companies were hiring a lot of juniors. This was...
Is this your brain on ChatGPT?
Is this your brain on ChatGPT? Read it on seangoedecke.com A recent MIT study - titled “Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an...
Every service should have a killswitch
Every service should have a killswitch Read it on seangoedecke.com The more time you spend designing systems, the more paranoid you get about things going...
Say "but yes", not "yes but"
Say "but yes", not "yes but" Read it on seangoedecke.com When you’re agreeing with someone but you have a caveat, don’t say “yes, but”. Instead, say “but...
The first big AI disaster is yet to happen
The first big AI disaster is yet to happen Read it on seangoedecke.com The first public passenger locomotive, Locomotion No. 1, began service in September...
The illusion of "The Illusion of Thinking"
The illusion of "The Illusion of Thinking" Read it on seangoedecke.com Very recently (early June 2025), Apple released a paper called The Illusion of...
AI interpretability is further along than I thought
AI interpretability is further along than I thought Read it on seangoedecke.com It’s common to call AI language models “black boxes”. Unlike normal human-...
Arguing point-by-point considered harmful
Arguing point-by-point considered harmful Read it on seangoedecke.com Engineers love to have technical discussions point-by-point: replying to every idea in...
John Henry and the large language model
John Henry and the large language model Read it on seangoedecke.com John Henry told his captain A man ain’t nothing but a man But before I let your steam...
Why DeepSeek is cheap at scale but expensive to run locally
Why DeepSeek is cheap at scale but expensive to run locally Read it on seangoedecke.com Why is DeepSeek-V3 supposedly fast and cheap to serve at scale, but...
It's not your codebase
It's not your codebase Read it on seangoedecke.com Engineers often feel a lot of ownership over the code they write. That’s understandable: it’s partially...
How does ChatGPT work? What is AI, really?
How does ChatGPT work? What is AI, really? Read it on seangoedecke.com How does AI work? Since 2022, it’s been possible to have a conversation with a...
How projects fail at large tech companies
How projects fail at large tech companies Read the original post at seangoedecke.com How do projects fail at large tech companies? As I've said many times,...
In defense of shallow technical knowledge
In defense of shallow technical knowledge Read it on seangoedecke.com Whenever a new piece of technology comes out (these days, mostly AI) I go to some...
Practical AI techniques for daily engineering work
Practical AI techniques for daily engineering work Read it on seangoedecke.com Popular advice about using AI is obsessed with tools and “magic prompts”. To...
Strengths and limitations of diffusion language models
Strengths and limitations of diffusion language models Read it on seangoedecke.com Google recently released Gemini Diffusion, which is impressing everyone...