Dear Reader,

Over the last few days, I showed you what AI agents can do for your research, how the system is built from five simple components, and how it saved my PhD defence. But does that work for you?

Before building the AI Agents for Academics Course, I surveyed 73 academics on their hopes for AI and designed the course to meet these goals.

First Question: What would you like AI to handle?

The "Other" category had more unique answers like "preparing teaching material" or "Automating Obsidian or Zotero"

How the course solves it:

• Literature Search & Discovery are done by having AI analyse your notes or papers and use the Consensus MCP to autonomously search for relevant papers (Lecture 1.8 and 2.9).
• For connecting notes and ideas, the course features a connector skill, that analyses a paper and finds similarities to other notes/papers for you to make sense of it (Lecture 2.8).
• Organising notes is done using the "custodian skill", which enforces a set of predefined rules across your vault (Lecture 2.11).
• Extraction of ideas from papers is done via a scoring system skill that compares papers "objectively" against predefined criteria (Lecture 2.10).
• For academic writing, I ran an entire webinar, and you can download the skills I used for outlining my PhD thesis and helping me to write it here.

This questionnaire greatly helped design the course. Thank you to all participants!

Questions 2 & 3: What do attendees want from the course & who are they?

To my surprise, the majority of participants were quite senior academics, and almost everyone wanted some sort of working system that would translate beyond my examples. (The other categories were people who described more how the course should be i.e. slow-paced, or geared towards humanities).

For that reason, I did two things:

1. Many of the skills are not just downloads, but questionnaires or scripts that you can run to design your own output. A little bit like Lego. The parts are predefined, and you can get a construction plan, but there is a lot of room for modification and experiments.
2. The founding cohort for those who join us in the next few days. I think having a group of people working on the same goal is exactly what you need to achieve your goals.

Time left to secure a spot in the founding cohort of the course:

​

The course is still available afterwards, but the cohort is closed.

Previous parts of this series​
Part 1: My PhD Defence: 6 days to rebut 30 pages of critique​
Part 2: What can AI agents do and how they helped me ace my PhD defence (Video)
Part 3: How to set up an AI agent system in Obsidian (Video)

This email is part of a 4-day series on AI agents. If you don't want to know about it, unsubscribe from this series.​
Want to leave forever: Unsubscribe from everything (never hear from me again)