Heidi is frustratingly consistent at hallucinating stuff. I've seen it in almost all of the dozen or so summaries I've had from medical people recently (surgeon, physio, consultant). A GP I know tried for a month and then was like 'it's not worth the risk exposure to me or my patients'.
I know at least one GP that has stopped using Heidi Health for transcription. He (and as I've noticed with transcriptions from my medical professionals) has noticed many errors, far too many to be comfortable. Things might improve, but not yet.
This is where I'm at as a GP. Every few months I give Heidi another try, but I haven't noticed any real improvement over the last two years. It spends lots of words on trivial nonsense and misses clinically significant points and sometimes entire issues. It takes far more time to review and fix the notes than it saves in typing. Presumably it will be good enough one day, but it's not there yet.
Git would just say untracked file. I couldn't add it. Finally chatgpt explained. Kosovo's airport code (PRN) is a reserved file/folder name on Windows (facepalm).
easiest one to get going with is to add the Playwright MCP. As a python dev you might have used it to do test automation? Anyway, it gives your tool eg Cursor, Claude Code access to the browser and automation using playwright. Meaning it can literally load a page to confirm its own change just had the desired effect.
The blender one is also fun as a starting point, if you do any 3d modelling (or even if you don't).
So what about flight searches where we have to query several 3rd party providers, and can take 45 seconds to get results from all of them (out of my control). I can dynamically update the page (and do) but a scraper would have to wait 20-45 seconds to get the 'cheapest' flight from my site. I can add async the queries and have the fastest pipes, but if the upstream providers take their time (they need to query their GDSs as well), there's not much you can do.
Oof, reminds me of the data pipelines I maintain that pull data from Salesforce. I've optimized the heck out of all of our internal database queries and they're blazing fast. Then my pipeline has to wait patiently for the SOQL response.
Basically a test of putting guard rails around format and content of a website and seeing how much I could automatically generate on a topic of interest to myself.
Biggest benefit I've seen with cursor is to write tests for everything. Far too mcuh content hallucination, or made up links at first, but once you put in some test guardrails you can minimise this.
I played with Kepler GL a couple of years ago to visualise the Canterbury quakes. It's kinda phenomenal to see how frequent the little ones are, and the aftershocks from big ones take a while to drop off...
Heidi is frustratingly consistent at hallucinating stuff. I've seen it in almost all of the dozen or so summaries I've had from medical people recently (surgeon, physio, consultant). A GP I know tried for a month and then was like 'it's not worth the risk exposure to me or my patients'.
reply