I started testing AI tools out of curiosity, but I kept using them because they were useful in one very specific way: they removed drag from the parts of work that are necessary but repetitive.
The tools only stay in my workflow if they save time and make the output easier for other people to use.
Where AI helps me most
The highest-return use cases have all been documentation-adjacent.
| Use case | What AI does well | What I still do manually |
|---|---|---|
| Meeting synthesis | Extracts themes and action items | Validate nuance and ownership |
| Status report first draft | Builds structure quickly | Rewrite for audience and tone |
| Process documentation | Turns notes into a clear outline | Check for operational accuracy |
| Content repurposing | Adapts one idea into multiple formats | Choose the final narrative |
The line is simple: AI accelerates assembly, but I keep responsibility for judgment.
What I stopped using it for
I do not rely on AI for stakeholder-sensitive messages, scope tradeoff decisions, or anything that needs social context. The draft may look polished while still missing the one implication that actually matters.
That is why I see AI as a multiplier, not an owner.
The workflow that stuck
- Capture rough notes fast
- Use AI to structure the material
- Rewrite for the real audience
- Sanity-check dates, names, owners, and actions
That pattern works because it starts with my context and ends with my review.
What makes the output better instead of just faster
The trap is asking AI to finish the work. The better pattern is asking it to make the work easier to finish well.
| Keep | Avoid |
|---|---|
| Outlines, restructuring, summaries | Blindly sending first drafts |
| Drafting repetitive sections | Letting tone default to generic |
| Turning raw notes into clean prose | Treating generated facts as verified |
Written by
Damini Aswal
AI-Native Project Manager
Google Certified Project Manager focused on delivery systems, process clarity, and AI-integrated workflows.
Continue Reading
Career
Building a Blog That Thinks in Components
A sample MDX post showing custom callouts, styled images, markdown tables, and a reusable comparison table.
Agile
100% Sprint Delivery Isn't Luck — Here's the System Behind It
Six consecutive sprints with 100% commitment delivery. It did not happen by accident. It happened because of one daily habit and two non-negotiable rules.
Change Management
The Change Management Mistake Most PMs Make
Most change management plans focus on the system being changed. The ones that fail do so because they ignored the people using it. Here is what I learned.
Share This Post