Tony Alicea

10 Dec 2025

I'm pleased to present, for free, the first hour of my new course Understanding AI-Assisted Software Development on YouTube! This course is designed to be agnostic to any one AI tool, and instead provide a mental model and best practices for you and your team. 👇

We programmers have been abstracting ourselves further and further away from the "real code" for decades now. Memory management used to be a skill until modern languages and garbage collection.

Thanks to vibe coding, all those unnecessary UIs in movies that you couldn't have imagined a real coder taking the time to make...are now totally believable.

anthropic.comRedesigning source control at AnthropicAs our merges per day grew more than tenfold, source control became our biggest development bottleneck. Here, we share how we migrated to an in-house version control system in five months.

2027 will be the era of agentic software factories. Every other aspect of the creation of software will have to adapt. www.anthropic.com/engineering/...

tonyalicea.devFalling In Love With the Problem, Not the Solution, Is More Important Than Ever | Tony AliceaTyping code and dragging pixels were always solutions. What problem were you trying to solve?

First new blog post in awhile. It's a short one, but a response to a lot of conversations I've been having with devs and designers who are unsettled and unhappy as AI changes the industry. tonyalicea.dev/blog/fall-in...

22 Sept

Why do some of the new Alexa voices have vocal fry? Why?

16 Sept

AI for developers with imposter syndrome is like drugs for people with addictive personalities.

15 Sept

Introducing Sloptris! A Tetris clone that satires what AI-generated coding is like. I've felt a lot like AI-assisted dev shares elements of game experiences. This is a fun little idea I had to express the experience within a game. 👇

tonyalicea.devSloptris - if Tetris was like AI-assisted coding

15 Sept

Gameplay: - Build blocks manually or "AI-generate" them - AI tetriminos might mutate after dropping - Spend valuable time reviewing AI tetriminos to know how they'll mutate Try it out (beta): tonyalicea.dev/sloptris/

9 Sept

I've been so pleased to see the reaction to the UX.md concept in the design and product community. I'm happy to share an interactive report of my own small experiments with how UX-context impacts product features that AI generates. 👇

tonyalicea.devUX.md playground | How context changes AI outputCompare sixteen AI-generated insurance claim pages to see how research findings, user and world models, and a glossary affect the design.

9 Sept

Check out the interactive report here: tonyalicea.dev/ux-context/ There's a lot more serious research and training coming from @nngroupux.bsky.social on AI topics - stay tuned!

1 Sept

There's a lot of possibility for using AI to build great software. But the simple act of using AI to build software is no guarantee of greatness.

1 Sept

We are confusing the ability to build faster with the ability to build better.

24 Aug

The most popular AI loop seems to be: 1. Human: "I'm bad at writing." 2. <Asks AI to write for them> 3. <AI output is more difficult to read than what the human would have written.> 4. Human posts the AI slop. 5. Human gets worse at writing due to lack of practice. 6. Go to 1.

19 Aug

Stop telling me about how many PRs you shipped. Faster implementation doesn’t mean faster thinking. If anything, it questions how much thought went into what you’re doing.

12 Aug

"How fast we can build" and "how fast we can think" are not the same thing. It is a skill and cultural regression for product teams to conflate the two.

6 Aug

With every new iteration of tooling and advancement in software development, the hope of organizations seems to be that we can skip that pesky, time-consuming part called thinking. That's never true.

youtube.comDesigning Filters: A 4-Step Framework | NN/G LIVEYouTube video by NNgroup

29 July

Free NN/G webinar starting up on YouTube in an hour. Designing filters is a deeper problem then you may realize, and you'll get a framework to help you make great, usable decisions. I'll be moderating the chat, highly recommended! www.youtube.com/live/BfVWcom...

27 July

What is the role of product/UX design and user research in an agentic workflow? In my first ever write-up for @nngroupux.bsky.social I'm excited to discuss UX-Context Design. Is it the future of UX design and user research? 👇

nngroup.comUX-Context Design: Using UX Knowledge to Inform AI-Generated DesignAs more interface work is AI-generated, the output of research and design shifts from documents written for humans to curated context that guides AI.

27 July

Read the full write-up: www.nngroup.com/articles/ux-...

github.comGitHub - AnthonyPAlicea/skills: Skills for Devs Who Want to UnderstandSkills for Devs Who Want to Understand. Contribute to AnthonyPAlicea/skills development by creating an account on GitHub.

16 July

A reminder that my free agent skills are now available, all designed to help you avoid skill atrophy and maintain accountability for you and your team as you engage in AI-assisted development. Don't imitate, understand! github.com/AnthonyPAlic...

gameswithtony.comGORILLAS: RepeeledA browser tribute to QBasic GORILLA.BAS — wide destructible city, dynamic camera, hotseat or vs. AI.

16 July

If you want to experiment with AI-assisted dev, do fun side projects! This is my snazzed up version of QBasic Gorillas, a game I played a ton with friends back in the day. Anyone else remember it? gameswithtony.com/gorillas/

13 July

This idea that you can now constantly change software based on 'feedback' with no up-front research assumes that users will stick around while you keep changing software under their feet. If people have to constantly re-learn your software, what's keeping them from moving on?

18words.com18 Words - Daily Word ChallengeA fast daily word challenge. Can you survive all 18 words?

10 July

I'm addicted. 18words.com

8 July

I used Fable to help me build a daily puzzle game inspired by one of my favorite board games (Vital Lacerda's On Mars). Check out Oppy (in beta): gameswithtony.com/oppy

8 July

It's extraordinary to me to watch professional people use AI to generate a three sentence comment in their native language. It takes more effort to prompt the AI to write the comment than it does to write it yourself. And the comment is obviously AI-written. It's sad.

1 July

There's a lot of devs who are trying to become product engineers and forward-deployed engineers and will be sent to go talk to users...and have no idea that talking to users, getting them to open up, and not biasing them is a learned skill. For example, contextual inquiry.

1 July

Announcing my new agent skill: /de-div It cleans up div soup and helps you author semantic, accessible HTML, working with you when a semantic choice is unclear and generating a report for you to learn from at the end. 👇

1 July

Repo: github.com/anthonypalic... Install: npx skills add github.com/anthonypalic... --skill de-div GitHub stars appreciated!

30 June

If you're familiar with the board game "On Mars" by Vital Lacerda, I've made a daily puzzle homage to it called Oppy. It's in beta, check it out: gameswithtony.com/oppy

25 June

Got this nice comment on my new do-i-understand agent skill: "Thanks for this skill — it's genuinely useful, and we got real value from it while adopting it in one of our repos." Running this skill benefits your knowledge and your app security. github.com/AnthonyPAlic...

23 June

Claude's current favorite words seem to be "gap" and "quietly". I start zoning out whenever I read them now.

18 June

Remember when "slop" meant bad food?

I just released my first agent skill: /do-i-understand As a technical educator for many years, I want to maintain my own skill and that of those around me. This agent skill makes sure you understand what you're shipping before you merge it. Here's how it works 👇

Third, for open source maintainers. You can ask for a do-i-understand report in PRs as a light honor-system check. It's a bit of intentional friction before a merge request. Free and open, link below. I really enjoy using it, so let me know what you think.

Find /do-i-understand here: github.com/AnthonyPAlic...

Three ways to use it: First, on yourself. It keeps understanding sharp and helps you be accountable for the code you're merging.

Second, with junior engineers. Require it on their PRs, but build the safe space first. "I don't know" has to be a welcome answer, one that invites pair coding and mentorship instead of something to hide.

Run /do-i-understand and it looks at the change you're about to push (or point it at earlier commits). It prioritizes changes, then probes your reasoning, your assumptions, and the ways the change could fail. It's looking for the parts you don't yet fully understand.

At the end it writes a short report. Paste it into your PR description for your reviewer, or keep it as a personal record. The goal is simple: *push code only once you can be accountable for it*.

5 June

"Here's the thing..." <short pithy sentence> "Quietly..." "Nobody..."

5 June

If your process is now to make feature requests into tickets and have your agent constantly turn tickets into PRs...with no triage of those feature requests or root cause analysis...you are going to end up with terrible, terrible software.

nngroup.comNielsen Norman Group: UX Training, Consulting, & ResearchA leader in the user experience field, NN/g conducts groundbreaking research, trains and certifies UX practitioners, and provides UX consulting to clients.

4 June

Devs: in the age of AI you need to differentiate with good user experience, know how to talk to your users, and build product taste and sense...to do that you should be getting familiar with Nielsen Norman Group (if you aren't already). Seriously: www.nngroup.com

0llm.tonyalicea.dev0llm — spend time away from AISpend some time without an LLM.

1 June

Zero LLM is an effort to take some time without an LLM. 1) Agree to some time without AI. Do things yourself. 2) Start a session to track your time. 3) When you feel the addictive itch to ask AI for help, go to the site instead. 4) End your session and share with #0llm.

31 May

I had this idea. We are using LLMs too much. We need a recalibration sometimes. A bit of time where you agree with yourself to do things manually, keep your skill and your voice. Check out 0llm. If you use it, share it with the hashtag #0llm.

0llm.tonyalicea.dev0llm — time in your own wordsSpend some time without an LLM. A quiet recalibration — type it yourself, in your own words.

31 May

Try it out here: 0llm.tonyalicea.dev

30 May

Making content is cheap now. The question is if you have something to say.

25 May

A problem with AI output is what we’ll call “effort bias”. Because an output (text, image, video, etc.) looks like things that have historically taken a lot of human effort and thinking, we assume a degree of correctness.

23 May

AI has caused a current trend of output over outcomes. That can only last so long.