One part of me loves this idea, while the other is apprehensive. This concern mainly stems from the possibility of capturing sensitive information, such as passwords, bank account websites etc. How is this being addressed?
Great! Love the idea and installed it. Something that's not quite obvious to me: How do you let your AI "know" about all of this? Skill, reference in AGENT.md?
Woo! In the onboarding, we let you install a skill, and you just select which AI you're using and it will place the skill in the right place for it to use it!
Hey Jatin, super valid concern, thanks for bringing it up.
From what we've seen ourselves and from early adopters, the power usage of Familiar is not significant. Im able to run hours with my macbook air m1 and Familiar.
Hey Lillian! It makes total sense, and it's still super early and an alpha testing. I'd love to hear what are some of the things you'd love to learn more about that would make you feel better about this. That would help us make sure we're focusing on the right things.
Will def respond via private email. I am so glad that Write Grow Sell introduced me to you, your ecosystem, and your great newsletter here on Substack!
As good practice, how often should we keep cleaning up the stills folder and how to clean up irrelevant data while keeping relevant stuff? I am worried to leave it on all the time since it may lead to data accumulation.
Thank you! That's a great question, and you're not alone. The latest version automatically cleans up your still images after two days. You have the option to extend that to seven days if you'd like to keep them for longer.
Definitely - we don't want people to download it directly. Ideally, we'd like for people to contact us to onboard at this stage. We'll launch a landing page when it's available for a general download. Of course, people can go seek it out, but we hope most people will wait
That's awesome, I'm glad you're taking such a safe approach to it. As long as everything remains local, that already puts my mind at ease compared to all the predecessors (that I tested). I'd love to test it out! Though, I'll need to circle back to this next week or so.
Tal! I'd love to experiment with it, though I am intuitively doubtful CC will not drown in all this context bloat.
Btw, unrelated - what do you use instead of granola when you have meetings in hebrew? I feel like a caveman taking manual notes (I still do it in granola though).
I realize you asked Tal, not me, but I'll point you in the right direction -- use the Wave.co notetaker. It can take the meeting notes in Hebrew (or English) and present the summary and transcript in Hebrew. It also does a pretty decent job in identifying different speakers in the same meeting, which Granola does not yet offer.
thanks for the comment! we have logic in place that filters out single words from the clipboard. any password or api key that you copy wont be saved :)
One part of me loves this idea, while the other is apprehensive. This concern mainly stems from the possibility of capturing sensitive information, such as passwords, bank account websites etc. How is this being addressed?
Definitely. Same! Familiar skips passwords, tokens, and API keys. Check that part out here: https://github.com/familiar-software/familiar/tree/main/src/security
To be extra safe, you can pause or quit Familiar anytime. You can also delete recent captures.
Great! Love the idea and installed it. Something that's not quite obvious to me: How do you let your AI "know" about all of this? Skill, reference in AGENT.md?
Woo! In the onboarding, we let you install a skill, and you just select which AI you're using and it will place the skill in the right place for it to use it!
https://github.com/familiar-software/familiar/blob/main/src/skills/familiar/SKILL.md
Hey Jatin, super valid concern, thanks for bringing it up.
From what we've seen ourselves and from early adopters, the power usage of Familiar is not significant. Im able to run hours with my macbook air m1 and Familiar.
Amazing! One of these days I'll be ready to test, but first need to learn a bit more to feel comfortable
Hey Lillian! It makes total sense, and it's still super early and an alpha testing. I'd love to hear what are some of the things you'd love to learn more about that would make you feel better about this. That would help us make sure we're focusing on the right things.
Will def respond via private email. I am so glad that Write Grow Sell introduced me to you, your ecosystem, and your great newsletter here on Substack!
Nice! Curious though: since it is always running in the background taking snaps at real-time, doesn't this drain too much battery of the mac?
This is awesome!
As good practice, how often should we keep cleaning up the stills folder and how to clean up irrelevant data while keeping relevant stuff? I am worried to leave it on all the time since it may lead to data accumulation.
Thank you! That's a great question, and you're not alone. The latest version automatically cleans up your still images after two days. You have the option to extend that to seven days if you'd like to keep them for longer.
Congrats on the launch!!
Already using this daily and beginning to integrate into my workflows as a always on memory of what I did
Mobile anytime soon?
Love this. Download can be simpler. Users can download the wrong architectural release. Opened a PR to help with this. https://github.com/familiar-software/familiar/pull/7
oh wow badass PR!!!!
Definitely - we don't want people to download it directly. Ideally, we'd like for people to contact us to onboard at this stage. We'll launch a landing page when it's available for a general download. Of course, people can go seek it out, but we hope most people will wait
That's awesome, I'm glad you're taking such a safe approach to it. As long as everything remains local, that already puts my mind at ease compared to all the predecessors (that I tested). I'd love to test it out! Though, I'll need to circle back to this next week or so.
Future gets closer every day! Couldn’t imagine you building anything less extraordinary and aspirational!! Looking forward to trying it out.
Tal! I'd love to experiment with it, though I am intuitively doubtful CC will not drown in all this context bloat.
Btw, unrelated - what do you use instead of granola when you have meetings in hebrew? I feel like a caveman taking manual notes (I still do it in granola though).
I've also had success with Quill Meetings for notes in Hebrew :) for anyone interested.
Grain! I was also doubtful, but I learned to let go and accept that these agents are now smarter than my brain can wrap my mind around.
I realize you asked Tal, not me, but I'll point you in the right direction -- use the Wave.co notetaker. It can take the meeting notes in Hebrew (or English) and present the summary and transcript in Hebrew. It also does a pretty decent job in identifying different speakers in the same meeting, which Granola does not yet offer.
Oooo interesting!
thanks for the comment! we have logic in place that filters out single words from the clipboard. any password or api key that you copy wont be saved :)