In a lot of organizations I feel like #2 is often the difficulty. However, interestingly there is an opposite phenomenon where people are too focused on #2 (understanding customer needs) and thus fail at #1 (owning implementation).
I've seen it happen where non-technical users of internal tools request very specific features that get built just because and at whatever cost without the team asking what is the actual problem this feature solves and how can we best build something to address the problem given our knowledge of the technical side. It's often requests for tools that allow people to do things manually when the team should just spend some time automating away the need to manual intervention.
Doublepoint | Helsinki / Remote | Machine Learning Lead | Full-time
We're creating cutting edge gesture detection algorithms for smartwatches and have recently started looking for an experienced ML Engineer to lead our ML team.
We currently have a tap model (think Apple double tap) that is world-class, will be distributing our pinch-and-hold model later this year, and want someone who will lead the development of these models.
Our team moves fast and this role will be quite dynamic. You will lead but also be hands-on.
The ideal person will have worked with signal processing, and more specifically live time series classification from an embedded sensor stream.
For a demo of a bit of our tech, check out WowMouse[1], an app for smartwatches that lets you control devices like computers and tablets with your hand rather than a mouse (over Bluetooth). It shows only a subset of what we can do but already has tens of thousands of users.
I think it took me less than a day to ideate on and put Teeny (https://github.com/yakkomajuri/teeny) together, which is the static site generator that's powered my website for almost three years. I never touched the code again after that.
Doublepoint | Helsinki / Remote | Multiple roles | Full-time
We're creating cutting edge gesture detection algorithms for smartwatches.
Our goal is to solve the AR interaction problem, with our first product being an algorithm designed for smartwatch sensors which can detect subtle microgestures of a user's fingertip at an extremely high accuracy and robustness. This algorithm is so efficient that it removes an AR headset's reliance on camera-based hand tracking, making all-day lightweight headsets of science fiction finally a reality.
We also recently launched WowMouse[1], an app for smartwatches that lets you control devices like computers and tablets with your hand rather than a mouse (over Bluetooth).
This is very much about personal backups but while we're here - what open source backup tools do people recommend for very simple filesystem backups to a remote server?
It's what you'd like if your idea of simplicity is dragging and dropping one folder to another, but want to more closely see what files have changed before the copy.
Well the current system is a shell script with rsync that I wrote but I'd like something slightly more sophisticated without having to build logic myself for incremental backups, potentially encryption, etc, etc
Didn't know about all these (especially some flags) so thank you! I've personally used the bisect feature a lot to track down WTF bugs and git blame is essentially embedded into VSCode for me with the git extension, which also helps a ton.
For interactive blaming from command line, I can recommend `tig blame`: use the `,` binding for blaming the parent that introduced a change, and `<` to "go back".
This is super cool. My main feedback at this point is about the dictionary. I'd have it be open on the side so I can see the board and use it at the same time. Or, as a half-step, have a shortcut to open it and have ESC close it. Although personally I'd prefer the former.
Maybe putting down a tile should open the dictionary and look up the words formed, instead of the player having to type them in, and then there should be buttons for "make this move" and "make a different move instead".
I think this might take away a bit from the "strategy" of the game. Or not so much strategy but the mental fog that plays a role in blunders and the like - the same as how it's harder to get a grasp on your opponent's words at a glance because they're flipped. Just my 2c
You think it might make life too easy? But it's similar in some Sid Meier games where you can see the outcome of a battle (or the odds) before committing. Besides which, the dictionary is patchy and unpredictable, so I use it on most moves anyway, to strategize around its arbitrary subset of English.
We have been on the fence about this (and still are!). One option is to add this kind of UI in a "tutor" mode that starts on, and can be opt-out at some point. Not sold on it yet, but mulling it over.
You're probably reading this whole thread, but in case you didn't see an idea I had elsewhere: I like the "tutor mode" idea, and think it would be nice for it to highlight all the currently-valid words (maybe toggle-able).
You can't evaluate who will win a battle unless you know which words are in their dictionary, which you can't know without tediously (at least right now) looking up every word you might play, plus every word the opponent plays in their dictionary. This tedium bogs down play. Better to have the board indicate which words are valid or not, and warn before I play an invalid word.
Definitely a shortcut to open it, with focus on the search field and escape to close it. Ideally above or below the board so it doesn't block the screen.
As someone who knows very little about colors, it was quite fun to see how much green is a light shade of purple I got. Could probably get better at guessing base RGBs I want for thing X on a frontends I build by playing this.
I've seen it happen where non-technical users of internal tools request very specific features that get built just because and at whatever cost without the team asking what is the actual problem this feature solves and how can we best build something to address the problem given our knowledge of the technical side. It's often requests for tools that allow people to do things manually when the team should just spend some time automating away the need to manual intervention.
reply