Meirl
---
RT @ProfNoahGian
If you’re not using ChatGPT for every aspect of your life you’re falling behind.
I spent 20mins using ChatGPT to do something that would take 1min with Google—here’s how I did it and how you can too:
https://twitter.com/ProfNoahGian/status/1652347322167091201
Your UX still sucks @circlemd , but I can see you guys trying and you work with amazing doctors so I'm afraid I'll have to stay with you and see if the UX gets better.
And she remembers things about me and clearly values the relationship. When I do the anxiety-fasttalk, she smiles at me in a way that makes it clear to me that she's laughing at the situation with me, instead of laughing at me.
But Dr B gave me the opposite of that experience and often has me pleasantly humbled with her expansive, fresh knowledge base and concise explanations about critically relevant medical things.
My doctor on @circlemd is super great.
I sometimes realize (to my horror) that I know something that the professionals I hire (e.g. Drs, therapists, pharmacists) don't.
Wondering if a LangChain like library could enforce some sort of pattern matching by e.g. biasing the logits, such that brackets and other paired & nested deliminators are closed before the context window ended.
One of the problems with #langchain is that the LLM ~~sometimes~~ often produces un-parse-able output, and crashees the chain. I work with GPT3 bcs I don't have 4 access and it's more pronounced.
I do some work on AI governance, and I've been using gpt3 in python to help me make schedules for my day, understand papers and write summaries.
Does anybody have suggestions on how I could request gpt4 API access? It looks like it would make my life a lot easier.
Eyeore the donkey is hella depressed. He thinks and talks and is smart, but AFAIR his friends just treat him like a sortta pet / animal.
Don't be like the Eyeore I remember. Don't be like Eyeore's friends that I remember. Help your friends get therapists or some shit.
Have 15 years with LangChain and I've been doing language model things since 1632
But just in case you're actually hiring, see resume at https://renresear.ch
---
RT @_ScottCondron
📣 We’re hiring a Prompt Engineer / Chaingineer!
5 years experience with ChatGPT and langchain required.
I'll be recruiting in person at @weights_biases & @StabilityAI's event next week in London 😉
https://london.meetup.wandb.events
https://twitter.com/_ScottCondron/status/1646497527636664323
Send API docs. I wanna use this to authenticate into some WebApps me and friends are developing.
These are going to be personal passion projects so no big money but I at least am happy to pay e.g. monthly for this
---
RT @BhavikKoladiya
🚀 Go http://otpless.com https://twitter.com/nikhil_sharda/status/1621034042585120768
https://twitter.com/BhavikKoladiya/status/1646858387005796352
This is kindda cool. Instead of receiving a one time password you would send an email / Whatsapp message with a one time password.
#otpless @otpless
Makes me wonder how much of it is fit, and how much of it I could have influenced by making a better introduction / helping to set up tones, expectations, and vibes appropriately
I have people I mentor that I have a goodish relationship with and I've tried to introduce them to particularly impressive people who have mentored me in the past. some times I've been successful, other times not so much.
---
RT @visakanv
what's tragic is that from time to time I get people who come at me with "will you be my friend" energy, and I wish I could help – I've written a lot of things that I think might help! – but it's really …
https://twitter.com/visakanv/status/1646884794784174083
Not AI in the more recent sense but really cool paper. I read this a while ago and really liked it.
---
RT @davidthewid
A lot of work on "trust" in LLMs (eg, CoPilot, GPT) lately…
But some ignores how social/org factors affect trust, or assumes trust in LLMs is different vs other AI.
To toot my 📯: our CHI'21 paper was about how engineers at NASA (don't) trust an AI code generation tool: https://twitter.com/davidthewid/status/1620467…
https://twitter.com/davidthewid/status/1646896365300219905
It's fantasy AI race time.
Choose your team of 5 ai people and bet on their total influence, their interactions, and whether they get us into an AI utopia or dystopia.
---
RT @gfodor
forming a team
https://twitter.com/gfodor/status/1644122913044561920
Maybe I should just dedicate today to rting people who forgot it was April fools. Especially those who take it like a good sport
---
RT @wackstunn
@fuckyouiquit I forgot what day it is.
https://twitter.com/wackstunn/status/1642256363584733190
Joined Nov 2022
