haystack-workshop-2025

Aryn Haystack 2025 Workshop

Setup

Install uv if you haven't already. On mac that's HOMEBREW_NO_UPDATE=1 brew install uv. We'll use uv to manage the virtualenv for this project.
Install poppler. Instructions (and reason why we need it) are here. Poppler is a PDF manipulation library we use to render PDFs as images, mostly for display purposes.
Clone this repository git clone git@github.com:aryn-ai/haystack-workshop-2025.git && cd haystack-workshop-2025
Pull the sycamore submodule git submodule update --init. This pulls down the sycamore source code, which means I don't have to make a release with every tweak while developing the workshop.
Set up python venv with uv uv sync
Download required data with make downloads. This includes 92 earnings call transcripts and a materialize directory (more on that in the tutorial)
Get an Aryn api key and an OpenAI api key. Export them as environment variables:

export ARYN_API_KEY="<key>"
export OPENAI_API_KEY="<key>"

Name		Name	Last commit message	Last commit date
Latest commit History 28 Commits
scripts		scripts
sycamore @ 414506c		sycamore @ 414506c
.gitignore		.gitignore
.gitmodules		.gitmodules
.python-version		.python-version
Earnings_Call_Ingestion_Script.ipynb		Earnings_Call_Ingestion_Script.ipynb
LICENSE		LICENSE
Makefile		Makefile
QuestionAnsweringNotebook.ipynb		QuestionAnsweringNotebook.ipynb
README.md		README.md
Rag.ipynb		Rag.ipynb
error_messages.py		error_messages.py
ingestion_tutorial.ipynb		ingestion_tutorial.ipynb
pyproject.toml		pyproject.toml
qa_tutorial.ipynb		qa_tutorial.ipynb
s3_download_list.txt		s3_download_list.txt
s3_download_list_http.txt		s3_download_list_http.txt
uv.lock		uv.lock
workshop_nb_1.ipynb		workshop_nb_1.ipynb
workshop_nb_2.ipynb		workshop_nb_2.ipynb
workshop_nb_3.ipynb		workshop_nb_3.ipynb
workshop_nb_4.ipynb		workshop_nb_4.ipynb