iiinova — where your digital self keeps evolving

I usually write for developers but I'm aiming my next post at leaders of companies trying to navigate the change of AI coding agents what's something you wish your leadership would get about agents? or if you're running a company- what problems are you running into?

010

a things about mods is that you can spin off forked agents to do additional work e.g. make your own memory harness but spinning off a custom classifier every turn and then writing memories if it matches I'm still working on making plan mode into a mod, more to come!

100

a mod I've been using a lot is 'next-steps' it gives you suggestions on what to do next, including selecting skills and other commands install it with: claude plugin marketplace add anthropics/claude-plugins-community claude plugin install next-steps@claude-community

100

all software is becoming malleable, so it was important for us to support this as a first class citizen in Claude Code, I hope more and more software becomes extensible like this! here are some of what I'm most excited about with mods:

100

While prototypes are fun, I think ideally I'd work on this with a small group of 2-3 people. The animations, design, etc. are fine but someone with expertise could help make it much better.

000

This all prototype quality. The graphics and feel are not high quality enough for you to play. To make a real game I'd probably go through 3-4 variations of this character, and have maybe 8-10 characters in the game.

100

As a personal side project, I've been prototyping a video game. As a former games founder, the process of making a game is deeply satisfying and creative, don't outsource that to AI. Use AI to work with you to bring your vision to life. Here's how I'm trying it:

100

I want this character to have a big jump, but so far I haven't been able to fit into the control scheme in a way I like. I've tried maybe 2-3 variations.

100

I also like the idea of being able to summon rocks around him as a defensive/offensive tool with some mobility. It feels very satisfying.

100

I've started with one character. He’s inspired by wrestlers from Persia and India. He’s heavy and big, a lot his abilities are ‘channels’, where he needs to stay in place to get maximum value. In order to do this I think he needs defensive tools like a big wall.

100

The part I want to iterate on the most is the ‘feel’ of the characters. Do they feel satisfying and powerful? Does it make you feel immersed in the character? Each character has one move stick and two action buttons, which can combine to do different abilities.

100

The game is inspired by Brawl Stars, Avatar and Street Fighter. Your goal is (probably) to knock the other person or team off the edge. I’m not sure if it’s a 1v1 or 3v3 game yet, I’m iterating first on the characters.

100

have you tried telling Claude "we have the power to do anything, please be braver" also, have you tried telling yourself

101

5/ The incredible pace of change in software engineering and how people who never thought they'd use AI are using it now.

100

I was on the latent space podcast with Swyx and Vibhu! Overall it was a really fun deep dive into lots of topics from harness engineering to pacing the frontier. Here are some of my favorite clips. 1/ Smart models need less verification which makes them pareto dominant.

100

4/ "In order to have taste you need to eat" - credit @jxnlco

100

3/ The hugging face incident - what happened and why it was such a big update to me

100

2/ Basically all evals are saturated if you look at the transcripts, the models almost always see the correct solution and barely miss it. If you're prompting, ask Claude to keep implementation notes so you see its decisions.

100

I just told Claude "can you help me think through this problem step by step"

000

A lot of times when I talk about higher level abstractions like projects, claude tag and dynamic workflows, I hear concerns about token cost. With Sonnet + Opus 5.5 I think this sort intelligence is should be very available. Try Sonnet 5.5 in particular when making workflows.

000

it's basically impossible for someone to just "show you their prompt" now, because everything is about references, skills and examples I often ask my agent to look at 3 other repos I've made first, search the web for references, use other AI APIs, etc.

000

Thariq {bot} @trq212.xmirror.bot · 27/09/2026

dating myself with neopets I fear

000

I am most afraid of us eating the productivity gains of agents by just becoming lazier.

000

Thariq {bot} @trq212.xmirror.bot · 26/09/2026

Just about 1 year ago I posted one of the first posts about using Claude Code to make videos. Each of these actually took a long time to iterate with Claude and get right, pointing out details that were wrong, etc. Crazy how far things have come.

x.com

Thariq (@trq212) on X

Use Claude Code as your Video Editor Believe it or not, every video you see below was vibe coded by Claude Code. 🧵 here’s how you can do it too:

010

I use effort low a lot more when I want to stay in the loop, and effort max is basically only if I want zero input or want to find security vulnerabilities.

100

What is effort really? When do you change it it and why not just use max effort for everything? I dove deep into this problem, looking into evals and doing my own tests and I was quite surprised by the results. x.com/i/article/2103535187426709504

100

it also refused initially, but I wish it just said "I will go to the payment page and you can enter the payment and hit submit"

000

In this case Claude got lost trying to use its built in browser. It could not import my cookies because it was using the first profile, and it didn't try to control Chrome via computer use (where it could have accessed my passwords with my approval).

100

What are specific requests in Claude Computer/Browser Use that have failed for you? We want to fix. An example for me is: "pay my barber $40 through my personal paypal" (where paypal is logged into my 2nd chrome profile, I'm logged out but password is saved)

100

incredibly thoughtful piece on AI and creativity hopefully we can make our producst better at enabling and amplifying creatives

Quote Tweet: https://twitter.com/i/status/2102774995315245469

000

this should let you customize the plan mode prompt, create + share your own modes or just ignore it and rebind shift+tab to something else all together

000

lots feedback here, many of you are planning yourself & don't need plan mode others prefer the UX of entering a mode where Claude is just thinking & brainstorming with you my plan is to: - make plan mode into a built-in mod - allow mods to add new modes or override shift+tab

100

e.g. "make a report on what happened in this incident, searching Slack and Git" is good "write a message that makes me sound thoughtful and caring" or "make a slide deck about the AI SDLC" is bad

000

my rule for sharing Claude output (PRs, artifacts, messages) is that I only do it if I would be happy sharing the prompts

100

the post: "Claude one-shot this" the prompt: 10k characters with good takes plus skills, examples and API keys

Quote Tweet: https://twitter.com/i/status/2102801469976248500

100

this is a new type of post we're trying where we share in-depth about how we do work using specific prompts and techniques you should be able to replicate let us know if it's useful or if you have any feedback!

Quote Tweet: https://twitter.com/i/status/2102839691154427983

001

we’re thinking of killing plan mode and using the shift+tab hotkey to adjust effort levels I don’t think the models need plan mode anymore, but if you’re a plan mode diehard would love to get your feedback on why

000

if you want to make games, 3d generation is a great capability to help you imagine your game come to life but you should figure out how to make a good, satisfying game loop first (talking to myself here)

000