What are the key takeaways from this How I AI episode?

Key insights include: **Custom Task System:** Build personalized task management in Claude Code using markdown files with front matter (due dates, tags) stored in Obsidian, enabling natural language commands like "new task" instead of clicking through GUI interfaces daily.; **Automated Research Pipeline:** Set up Python cron jobs to search arXiv and Google Scholar daily based on custom keywords, automatically generate paper summaries overnight, and deliver filtered research digests integrated directly into your morning to-do list workflow.; **Context File Architecture:** Create small, focused context files (writing style guides, business profiles, product details) instead of one large file, allowing Claude to load only relevant context per task and deliver accurate, personalized responses without repetitive prompting.

What did Teresa Torres discuss on How I AI?

Teresa Torres demonstrates how she uses Claude Code to manage her entire workflow through custom task management, automated academic research digestion, structured context libraries, and AI-assisted writing without coding expertise. Key topics include: **Custom Task System:** Build personalized task management in Claude Code using markdown files with front matter (due dates, tags) stored in Obsidian, enabling natural language commands like "new task" instead of clicking through GUI interfaces daily.; **Automated Research Pipeline:** Set up Python cron jobs to search arXiv and Google Scholar daily based on custom keywords, automatically generate paper summaries overnight, and deliver filtered research digests integrated directly into your morning to-do list workflow..

How long is this episode of How I AI?

This episode is 43 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

How I AI

Claude Code for product managers: research, writing, context libraries, custom to-do system, and more | Teresa Torres

January 19, 2026

43 min episode · 2 min read

Teresa Torres

Episode

43 min

Read time

2 min

Topics

Productivity, Relationships, Leadership

AI-Generated Summary

Published Jan 19, 2026

Key Takeaways

✓Custom Task System: Build personalized task management in Claude Code using markdown files with front matter (due dates, tags) stored in Obsidian, enabling natural language commands like "new task" instead of clicking through GUI interfaces daily.
✓Automated Research Pipeline: Set up Python cron jobs to search arXiv and Google Scholar daily based on custom keywords, automatically generate paper summaries overnight, and deliver filtered research digests integrated directly into your morning to-do list workflow.
✓Context File Architecture: Create small, focused context files (writing style guides, business profiles, product details) instead of one large file, allowing Claude to load only relevant context per task and deliver accurate, personalized responses without repetitive prompting.
✓AI Writing Partnership: Use Claude as a writing reviewer rather than generator by maintaining detailed style guides with specific tone, audience, and formatting preferences, enabling real-time feedback, research assistance, and typo correction while preserving authentic voice.

What It Covers

Teresa Torres demonstrates how she uses Claude Code to manage her entire workflow through custom task management, automated academic research digestion, structured context libraries, and AI-assisted writing without coding expertise.

Key Questions Answered

•Custom Task System: Build personalized task management in Claude Code using markdown files with front matter (due dates, tags) stored in Obsidian, enabling natural language commands like "new task" instead of clicking through GUI interfaces daily.
•Automated Research Pipeline: Set up Python cron jobs to search arXiv and Google Scholar daily based on custom keywords, automatically generate paper summaries overnight, and deliver filtered research digests integrated directly into your morning to-do list workflow.
•Context File Architecture: Create small, focused context files (writing style guides, business profiles, product details) instead of one large file, allowing Claude to load only relevant context per task and deliver accurate, personalized responses without repetitive prompting.
•AI Writing Partnership: Use Claude as a writing reviewer rather than generator by maintaining detailed style guides with specific tone, audience, and formatting preferences, enabling real-time feedback, research assistance, and typo correction while preserving authentic voice.

Notable Moment

Torres identified a methodological flaw in a viral research paper about purchase intent before it spread widely, then published a critical analysis that became her top-performing LinkedIn post, enabled entirely by her automated research summary system.

Know someone who'd find this useful?