Daniel Kilders

 iOS guy at a tech company. Personal account. #buildinpublic portfolio of apps as a side-hustle 👋 @SwiftStudyApp @bottomline_app @gonorthlight

[Daniel Kilders @DanielKilders]

Holly shit, I’ve been quiet lately trying to get the hang of ✨marketing ✨

And today I’m Top 6 in the finance App Store 💥💥

Mamma mia

[Daniel Kilders @DanielKilders]

@Dimillian Honest doubts:
Why does the UI change so little within both modes and does the chat window have different capabilities?

The only difference I see is that it hides git

[Daniel Kilders @DanielKilders]

The new chatgpt app comes with a little agent dude that floats on your screen

[Daniel Kilders @DanielKilders]

@thekitze I honestly don't understand what the differences are between both

[Daniel Kilders @DanielKilders]

Am I the only one that had to completely donwload and reinstall codex again, to get access to Sol and the new UI?

Anyways, here we go 🚀

[Daniel Kilders @DanielKilders]

@robj3d3 @AnthropicAI You don’t actually lose access. I already cancelled and it works fine.

Sneaky wording.

[Daniel Kilders @DanielKilders]

@theo Will we get a reset tho… a bit useless to get it extended if everybody is maxed out

[Daniel Kilders @DanielKilders]

Sure, let's say they all do. We are still at 14 days sick leave. Unless no one is ever sick and germans only ever take burnout leave, this won't lower the number much.

Most companies account for around 8 sick days a year.
Germany ain't fixing the energy crisis, bureocracy and old AF industry with 6 potential days more of work

[Claude @claudeai]

Fable 5 is back.

[Daniel Kilders @DanielKilders]

@sobedominik @JuicyMacBattery Congrats!

How did you get it featured?

[Daniel Kilders @DanielKilders]

@plahteenlahti When you say “React Native vs native,” is RN counted as cross-platform (iOS+Android), or are you comparing iOS-only RN vs iOS native / Android-only RN vs Android native?

Cross-platform RN would naturally have a bigger revenue ceiling if compared against a single native app.

[Daniel Kilders @DanielKilders]

@steipete @openclaw No codex app? Only CLI?

[rahul @rahulgs]

yes things are changing fast, but also I see companies (even faang) way behind the frontier for no reason.

you are guaranteed to lose if you fall behind.

the no unforced-errors ai leader playbook:

For your team:
- use coding agents. give all engineers their pick of harnesses, models, background agents: Claude code, Cursor, Devin, with closed/open models. Hearing Meta engineers are forced to use Llama 4. Opus 4.5 is the baseline now.

- give your agents tools to ALL dev tooling: Linear, GitHub, Datadog, Sentry, any Internal tooling. If agents are being held back because of lack of context that’s your fault.

- invest in your codebase specific agent docs. stop saying “doesn’t do X well”. If that’s an issue, try better prompting, https://t.co/SOjpn47yxo, linting, and code rules. Tell it how you want things. Every manual edit you make is an opportunity for https://t.co/S1ZvtYQwta improvement

- invest in robust background agent infra - get a full development stack working on VM/sandboxes. yes it’s hard to set up but it will be worth it, your engineers can run multiple in parallel. Code review will be the bottleneck soon.

- figure out security issues. stop being risk averse and do what is needed to unblock access to tools.

in your product:

- always use the latest generation models in your features (move things off of last gen models asap, unless robust evals indicate otherwise). Requires changes every 1-2 weeks - eg: GitHub copilot mobile still offers code review with gpt 4.1 and Sonnet 3.5 @jaredpalmer. You are leaving money on the table by being on Sonnet 4, or gpt 4o

- Use embedding semantic search instead of fuzzy search. Any general embedding model will do better than Levenshtein / fuzzy heuristics.

- leave no form unfilled. use structured outputs and whatever context you have on the user to do a best-effort pre-fill

- allow unstructured inputs on all product surfaces - must accept freeform text and documents. Forms are dead.

- custom finetuning is dead. Stop wasting time on it. Frontier is moving too fast to invest 8 weeks into finetuning. Costs are dropping too quickly for price to matter. Better prompting will take you very far and this will only become more true as instruction following improves

- build evals to make quick model-upgrade decisions. they don’t need to be perfect but at least need to allow you to compare models relative to each other. most decisions become clear on a Pareto cost vs benchmark perf plot

- encourage all engineers to build with ai: build primitives to call models from all code bases / models: structured output, semantic similarity endpoints, sandbox code execution. etc

What else am I missing?

[Daniel Kilders @DanielKilders]

@juanjovn Is your experience with Claude that good?

I used to be on the 200$ plan for Claude code, but now I get consistently much better results with the 200$ for codex.

Speaking specifically for ios/swift development.

[Daniel Kilders @DanielKilders]

@juanjovn

[Daniel Kilders @DanielKilders]

@juanjovn El normal o con expansión ?

[Daniel Kilders @DanielKilders]

@alice_ercolani @matteo_spada Did you guys manage to find a fix for this ?

[Daniel Kilders @DanielKilders]

@dagorenouf Oooh that’s the grand seiko I got myself to remember my honey moon. Freaking beautiful watch. I would also choose grand seiko over Rolex in a heart beat.

Check out the new UFA they just released. Sweet af.

[Daniel Kilders @DanielKilders]

@dams_app What makes you think it’s shady?
You can achieve the exact same from blogs, ugc or whatever marketing channel that doesn’t directly link the app and forces the user to search for the term.

Not saying this specific one isn’t shady, but it’s not the only way

[Daniel Kilders @DanielKilders]

@manchatz Buena suerte 😂