I do wonder what models it uses under the hood.

ChatGPT already knows more about me than Google did before LLMs, but would I switch to inferior models to preserve privacy? Hard tradeoff.

This looks great! What’s missing for me to switch from something like Wispr Flow is the ability to provide a dictionary for commonly mistaken words (name of your company, people, code libraries).

It has something called "Custom Words" which might be what you are describing. Haven't tested this feature yet properly.

So is this already in Handy or you are referring to a feature of the underlying models you are still not actively using?

This is already in Handy in Advanced > Custom Words.

There is also Post Processing where you can rerun the output through an LLM and refine it, which is the closest to what Wispr Flow is doing.

This can be found in the debug menu in the GUI (Cmd + Shift + D).

I dig that some models have an ability to say how sure they are of words. Manually entering a bunch of special words is ok, but I want to be able to review the output and see what words the model was less sure of, so I can go find out what I might need to add.

There’s a PR for this which will be pulled in soon enough, I can kick off a build of the PR if you want to download a pre release version

Okay so it's more directly text replacements

https://github.com/cjpais/Handy/actions/runs/21025848728

There is also LLM post processing which can do this, and the built in dictionary feature

Stacked pull requests solve this.

It doesn’t matter how much you wait for review of a PR if you can just build on top of it.

Also, this workflow encourages smaller PRs that are faster and easier to review.

We implement this at my company through a tool called Graphite and review times usually don’t exceed 1-2 days.

Typefully | Full Stack Engineer | Remote (CET Preferred) | Full-Time

Typefully is an extremely polished web app for crafting, scheduling, and publishing exceptional content on social media, empowering over 150k creators and teams globally.

About us:

We’re a small (3), bootstrapped, dedicated team of skilled craftspeople passionate about building well-designed, polished, user-friendly software.

About the Role:

We’re looking for a Full Stack Engineer with a great product sense to join our team. If you join, you'll work on:

- Developing and refining new features, like deeper AI integration and real-time collaborative editing.

- Solving interesting technical challenges, including enhancing our powerful text editor.

- Co-owning the product, taking charge of your work from ideation to release and maintenance.

Apply here

https://typefully.com/jobs (Product Engineer)

Yes we do.

You can paste any plaintext in our editor and it will be split into tweets. Just put 4 newlines between one tweet and the next.

Cool. What about thread size limitations?

Also, it would be nice if the tweets were trimmed. I just pasted a bunch into your editor and they had an extra newline so all the tweets had whitespace padding at the beginning.

Hi Jonathan, Typefully is first and foremost a thread-writing app meant for publishing, this is why we need all those permissions.

Profiles could use a subset of permissions, but we prefer to keep things simple and have a unified login flow.

Readng is a great alternative to GoodReads: https://beta.readng.co/

It’s currently in private beta.

Amazing idea, just bought a copy.

Refreshing take. An actual analysis with threat modeling calling out all the fallacies of people that argue Whatsapp is better just because it uses E2EE by default.

Felice anno nuovo HN!

What a crazy year 2020 has been, looking forward to what's next.