~/talk

Going beyond ChatGPT: introduction to prompt engineering & LLMs

on YouTube 01h 00m
Talkat PyData Seattle 2023 · on Wednesday, 2023-04-26· at 15:30· Kodiak Theatre· [large-language-models, ai-assisted-development]

speaker

name:Ties de Kok
links:none on file

resources

video:https://youtu.be/CG3b6D3XNOE
slides:none on file

summary

Covers a concrete pipeline: uses OpenAI API (GPT-3.5/4) and LangChain in a Jupyter notebook on GitHub/Binder to classify app feedback by building a prompt template that returns JSON output and an index. Demonstrates it on 20 synthetic examples, shows token counting and cost math (~$0.004 for 20 items on GPT‑3.5), compares GPT‑4 quality versus cost, implements programmatic calls, warns about storing API keys, and shows a code‑execution demo that manipulates pandas — everything runs live in the notebook.

AI-written summary · about

chapters (10)

  1. 0:00Introduction
  2. 1:48Prompt engineering for data science
  3. 8:03Generative LLMs basics
  4. 18:09LLMs for feedback analysis
  5. 30:44Coding environment setup
  6. 32:35Performance issue classification with API
  7. 41:14Effective prompt design
  8. 51:15Edge cases in prompts
  9. 53:01API cost and tokens
  10. 58:57Scaling solutions

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.