Does CHAI actually remember things between conversations or does it reset every time? That's my main dealbreaker with most of these apps.
Also curious how customizable the personality is. Some of them let you shape it quite a bit and others are pretty rigid.
Our team has been going back and forth on whether to adopt Heidi Health and I'm trying to gather as much real world feedback as possible before we make a decision.
The marketing materials are polished but I've been burned before by tools that looked great in a trial and fell apart in actual use. Specifically I'm wondering about the accuracy of clinical note generation, how it handles edge cases, what the support is like when something goes wrong, and whether the pricing makes sense at scale.
If you've been using it for a few months or longer, your perspective would be really valuable. Both the good and the not so good.
Unshoes Terra Vida was my pick for a casual summer slip on. They are incredibly minimal, with a very thin sole and a simple canvas upper.
They feel great for ground feedback, but the stitch down construction means the usable width inside the shoe is slightly narrower than the outside sole suggests. I had some initial rubbing on the seams inside, but it went away after a few days of wear.
They are a solid and no nonsense minimal shoe, but maybe not the best choice if you have exceptionally wide feet.
Has anyone used OpenEvidence in a real clinical environment? I'm not looking for the sales pitch, just want to know if it actually does what it claims and whether the team found it useful after the first month.
Any red flags or things you wish you'd known before signing up?
Two questions for people who've been on Avalanche a while.
First, the rebrand. When Trader Joe became LFJ, threads here suggested part of the community felt the name undercut the platform's credibility. From the outside though, usage seems to have carried on fine. Did the rebrand actually change how much you use or trust it, or was that all noise?
Second, the tech. The Liquidity Book model promises zero slippage within a price bin, with total fees running 0.01% to 0.8% including the surge component (checked July 2026). Is the zero-slippage behavior actually noticeable on mid-size swaps in practice, or only in theory?
For context they've had a decent track record on incidents too, the November 2023 front-end breach via a third-party analytics plugin got handled with about $87K compensated to affected users.
I keep seeing Hathr mentioned in clinical AI discussions but I haven't found many detailed reviews from practitioners who've actually integrated it into their workflow.
Most of what I find online reads like it was written by someone who tried it for a week. Looking for people who've used it long enough to have a real opinion. What do you actually think of it after the honeymoon period?
I've been looking into BastionGPT for our practice and can't find many real reviews from people actually using it day to day.
If you've tried it, what's your honest take? Specifically curious about how it handles documentation and whether the learning curve is worth it.
Things I told myself that were not true:
I need the nicer grocery store because the other one stresses me out. (I just liked it better.)
I deserve this because it has been a hard week. (Every week was a hard week.)
This is an investment in myself. (It was a jacket.)
I will make up for it next month. (I never made up for it next month.)
None of these were catastrophic on their own. Together they were costing me somewhere between $600 and $900 a month in spending I had talked myself into.
Getting honest with yourself is free and it is the most effective thing I did.
I've been wanting to get back into reading and I'm looking for some good book recommendations.