In Part 1, we introduced agents as a new kind of digital teammate for higher education: AI systems that can understand a goal, use tools, follow a workflow, and help people complete multi-step work.
Our university is interested in engaging the assistance of someone in building out our agents. Could you refer us to someone who we maybe able to engage?
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat via Zoom or telephone to discuss further?
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat via Zoom or telephone to discuss further?
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat to discuss further?
This is one of the clearest frameworks I've seen for building AI agents. The biggest takeaway is that successful agents aren't defined by the model, they're defined by the workflow. Start with one job, connect trusted knowledge, add the right tools, build in human review, and continuously evaluate performance. That's how you move from an impressive demo to a system people actually trust and use.
Which of these building blocks do you think organizations underestimate the most: workflows, guardrails, evaluation, or ownership?
I think we're entering a phase where execution is becoming cheap, but judgment is becoming expensive. The better AI gets, the more valuable taste becomes.
The future of higher education isn't just using AI—it's designing agents that help students, faculty, and staff focus on what matters most: learning, research, and human connection.
So glad to see this. I’ve been building a system for Speaking skills. There are so many adults & young adults in need of communication skills. Especially in this renaissance period of technology. Impressed!👍🥰
Our university is interested in engaging the assistance of someone in building out our agents. Could you refer us to someone who we maybe able to engage?
I sent my email through DM. I look forward to connecting.
Hi Rujun,
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat via Zoom or telephone to discuss further?
Best,
Group Executive Chairman
University of the Commonwealth Caribbean (UCC)
wadams@ucc.edu.jm
Tel 876-899-5987
Can we communicate via email? wadams@ucc.edu.jm
I built our TA system - DM me I have it running.
Hi Paul,
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat via Zoom or telephone to discuss further?
Best,
Group Executive Chairman
University of the Commonwealth Caribbean (UCC)
wadams@ucc.edu.jm
Tel 876-899-5987
Hi there are you still looking to discuss? I sent you an email.
Cheers, =PAW=
Hello Paul; we would like to meet with you via Zoom to explore/discuss.
I can help! Dm me
Hi Nabila,
I'm from the University of the Commonwealth Caribbean (UCC).
You mentioned on Substack that you could help with building agents for higher education. We're currently exploring AI implementations across our programs and I'd like to learn more about how you might be able to support us.
Would you be open to a brief chat to discuss further?
Best,
Group Executive Chairman
University of the Commonwealth Caribbean (UCC)
Lets build agents for earth first🌍⚕️
My latest project is a Coaching Wizard to help w middle school students. Got a grant too.
Very interesting, so these agents can be built by just coding or they need to be built using platforms such as crewai, n8n, etc?
They can be created in both ways, by code or by no-code platforms.
Im using chatgpt to code a cannister agent in motoko.
This is one of the clearest frameworks I've seen for building AI agents. The biggest takeaway is that successful agents aren't defined by the model, they're defined by the workflow. Start with one job, connect trusted knowledge, add the right tools, build in human review, and continuously evaluate performance. That's how you move from an impressive demo to a system people actually trust and use.
Which of these building blocks do you think organizations underestimate the most: workflows, guardrails, evaluation, or ownership?
I would worry more about how to build AI agents with proper calibration, for any use.
I think we're entering a phase where execution is becoming cheap, but judgment is becoming expensive. The better AI gets, the more valuable taste becomes.
Check https://en.bitlogic.io/expertise/edtech/
I would switch this to…how to build a real business identity and human coherence, to actually make it work.
The future of higher education isn't just using AI—it's designing agents that help students, faculty, and staff focus on what matters most: learning, research, and human connection.
I like the note.
So glad to see this. I’ve been building a system for Speaking skills. There are so many adults & young adults in need of communication skills. Especially in this renaissance period of technology. Impressed!👍🥰
You guys are doing incredibly well.
agents are not good for learning.
middle men/women/machine muddle learning…
real brain needs real learning not agent induced system
AI awesome. But if you use it in your business and you make a mistake…. The risk can be jail time.
https://dtnadvisory.substack.com/p/the-safety-briefing-no-one-gave-you?r