> ping me whenever anyone with >10k followers tweets about clawputer
for this case, is that prompt + polling like OpenClaw/Hermes cron, or do you filter events on your side before waking the agent?
Mostly asking to understand the cost model for watches like this. Does the user pay for repeated agent runs, or only when a filtered match is delivered?
We at Qordinate have made a managed version - which works on Telegram, Slack, WhatsApp, and our own app at the moment.
With iMessage, Email, Teams in pipeline.
Think - if you don't want to do the hassle of managing this yourself, you can offload that to us, along with security aspects of prompt injection, and performance aspects of tool search, etc.
If you want to try, it's available for free right now since we are early at https://qordinate.ai
Have been following you guys and using Qordinate on and off, and I find it quite helpful. I use it for getting the daily summary of my notifications, emails, and news around my interests, and searching things across the apps now. Also, sometimes for maintaining expenses and reading lists. Waiting for proactive actions now, which I feel would move it more towards being an actual assistant that you guys have been pitching for
This is such a great tool. The ability to scrape the web for different types of articles and suggest the ones that i would be interested in reading based on my reading history is such a great feature. Really excited to see what more you guys come up with.
Thank you!
Yes, the next biggest piece we are working on is autonomous coordination between two Qordinate agents - your true personal assistant, which knows what it can share, with whom and in what context, autonomously.
What I like most is that it doesn’t feel like I’m talking to a corporate bot. The tone is very casual/real, but it still does the boring structured stuff well – recurring reminders, lists, follow-ups etc.
Yes! personality is the biggest piece we wanted to focus on apart from the capabilities of course - not very sycophantic like gpt, but not too rude as well. a balance, if you will.
Glad you liked it :)
Hey, I have been using this on whatsapp, and I got mindblown first time I used it, and really love its personality.
I use it for all sorts of reminders nowadays.
Only thing is I was facing some latex formatting issues last time I used but I don’t know maybe that’s a Whatsapp rendering issue
That is a known bug. Actually I haven’t even come around to fixing such issues yet.
Another thing you should probably (not!) try: Click Play on a paper and click on Next/Prev :) The only way you can recover from that is by refreshing the page. Sorry about these rough edges as of now.
I keep on bookmarking things and those get lost in the huge pile I already have. So I ended up doing this:
Created a Google form which takes two inputs: URL I want to save, and a comma separated list of tags.
Then when it's submitted, an apps script gets triggered which inserts the URL+list of tags as array into mongo dB.
Then I query for URLs containing atleast one of the tag of form submission, and email myself those URLs saying hey you previously have saved these related links, check if you wanna read them.
It helps me to easily query whenever I want to see things I've saved for a particular topic/tag
The term you are looking for is New Grad. A good point to learn early on - avoid region specific lingo, communicating clearly is as important as your coding skills.
If you haven't pursued your MS yet, it might be better to pursue it in your destination country at a recognized university.
There are much better supports for that into a career.
International education and experience is notoriously difficult to transfer to North America. Gaining your very first north american work experience is notoriously difficult too.
Also its helpful to learn what graduates are expected to know and how they're expected to be able to think and problem solve in your destination country. Do a skills and competency map of yourself vs destination market.
For example, In anyone's home country, thinking and problem solving might not be as important as taking instructions well depending on the needs of the industry (say there is a lot of outsourcing to build to provided spec..) compared to a destination country.
for this case, is that prompt + polling like OpenClaw/Hermes cron, or do you filter events on your side before waking the agent?
Mostly asking to understand the cost model for watches like this. Does the user pay for repeated agent runs, or only when a filtered match is delivered?