OpenTeams Engineering Blog
Python, AI & Open Source
Hands-on engineering articles from the OpenTeams team on Python tooling, AI agents, LLMs, and open-source workflows used in real production.
Latest Articles
Built in production, shared in the open.
01
PDF Table Extraction: Docling vs Marker vs LlamaParse Compared
A hands-on benchmark of three document-parsing pipelines on real-world PDFs — accuracy, speed, and where each one breaks down.
Read More →02
Sandboxing Code Mode for Local LLM Agents
How we let a local LLM agent run code safely, with isolation that keeps experiments contained without slowing the inner loop.
Read More →03
Plugin Playground AI Integration for Faster Plugin Prototyping
Wiring AI into a plugin playground so developers can prototype, test, and iterate on integrations in a fraction of the time.
Read More →04
Slow Down — Simple Lessons for Guiding AI and Shipping Better Code
The counterintuitive practices that make AI-assisted development faster in the long run: clearer intent, tighter feedback, better code.
Read More →05
We Benchmarked 6 Python Package Managers on a Real ML Project
pip, Poetry, PDM, Conda, and more — measured on a real machine-learning project. Here’s what we found on speed and reproducibility.
Read More →06
What I Learned Making a Local LLM Do Real Work
Notes from the trenches of running a local model on genuine tasks — the tuning, the trade-offs, and where local wins.
Read More →07
From Skill to Agent: When a Text File Isn’t Enough
The line where a static skill file gives way to a full agent — and how to know when your automation needs to cross it.
Read More →Building the Infrastructure for a Distributed AI Economy
We build owned intelligence in the open. Want the team behind these articles working on your stack?
Talk to Us