TECHNICAL · DELIVERY · LEAD ·
iOS · WEB · AUTOMATION · AI · DOCKER · REACT · NODE ·
PRODUCT ARCHITECTURE · TEAM LEADERSHIP · MANAGEMENT ·
0%
Thoughts
AI

Manus — How I Use AI to Build Presentations

A practical case study of using an AI agent to prepare client presentations in 10 minutes.

Manus is an AI agent that doesn't just generate text — it executes multi-step tasks: search for information, structure it, format it the way you need.

My typical case is preparing a client presentation for a new project. It used to take 3–4 hours: gather data about the niche, frame the problem, pitch an approach, format the slides. Now it's 40 minutes.

What it looks like in practice

I hand the agent context: who the client is, what the task is, what the constraints are. I ask it to build a structure. I get 7–10 slides with a “problem → solution → how we do it → why us” arc. I edit 20% — my own voice, the specifics of this particular project.

What works, what doesn't

Works well: structuring arguments, finding analogies and examples, phrasing slide copy. Doesn't work well: the specific context of your business — you have to supply it explicitly, the agent can't guess it.

An AI agent is not a replacement for thinking. It's a way to get from blank page to draft fast — and the jump from zero to 60% is the most painful one. The agent takes that jump for you.