Kuse 2.0 Starter Tutorial
By Kuse
Summary
## Key takeaways - **Upload diverse files for context**: You can upload various file types like documents, PDFs, spreadsheets, and images to provide context for your projects. [00:43] - **Summarize user feedback for insights**: Kuse can analyze feedback forms to surface key pain points and highlight actionable trends. [01:28], [01:42] - **Generate PRDs based on user insights**: Create product requirement documents by referencing summarized user pain points and existing product architecture. [01:52], [02:11] - **Create on-brand promotional visuals**: Generate promotional images that match the visual style of previous campaigns by providing past campaign visuals and relevant documents as input. [03:14], [03:44] - **Magic Pen for multi-format content**: Use Magic Pen to combine different files, like posters and images, and make edits directly on a canvas by drawing and typing instructions. [03:50], [04:33] - **Knowledge compounds with use**: Every file created in Kuse becomes context for future work, making the AI smarter and more effective the more it's used. [04:51], [04:56]
Topics Covered
- AI builds knowledge, not just documents.
- Summarize user feedback to inform product development.
- Generate PRDs and marketing materials from existing data.
- Instantly create on-brand promotional images.
- Edit images intuitively with a 'magic pen' tool.
Full Transcript
Hey everyone, welcome to QES 2.0. The
AIOS QE is a visual AI workspace where
your knowledge compounds over time. It
reads your material, understand your
goals, and help you to generate whatever
you need next. Now, let's walk through
how it works.
For example, I am a product manager of
QES working to launch a new video
generation feature. On the left, click
new project.
Name it.
And then here at the bottom, click this
one. And then I can upload as many
related files as I want like docs, PDFs,
spreadsheets, images or just anything.
Here I have the feedback form for users,
product architecture that explains the
current feature framework and also some
past product visuals. I will drag and
select everything and then upload them
to cues. Click open so that cues will
understand the contacts. And if you
don't have any files and just start by
shutting in the chat box, click ask
cues.
So before starting to craft the new
feature, I really want to understand
what users think about the video
generation function. I would click the
feedback file here
as an input and then ask Q's to
summarize the key parts of it.
Generate. Okay. So now Q's analyzed the
data surfaces the key pain point and
also highlight trends that we can
actually act on.
Now that Q's has summarized the user
pain points for me. So I really want to
create a video generation PRD based on
those insight. I can manually drag and
select
the key parts in the chat box and then
click quotes to add them as reference
for the PRD task. and then select the
product architecture file as another
input so that cues will build the new
feature based on the real framework
instead of generating unrelated plans
prompt
send
and then in second cues will build a
clean and structured doc let's see
wow then here is our first deliverable
Also notice that everything here is
editable as long as we click edit.
Then I can for example bold the sections
and then also add any new ideas right
here in the white space. Like I can type
in cues cues cues.
Next, I want to make a new promotional
poster for the new feature. And then I
want to make sure that it is aligned
with the previous campaign style.
So, I need to select an image from the
past campaign here and then also the PRD
that we just made as an input. To select
multiple files in the folder, you can
just press shift on the keyboard and
then click one by one. where you can
also drag and select all of the files.
I ask cues to create a promotional image
matching the visual files of the
selected one.
Send
and see Q's produces on brand images
instantly.
If you want to combine different files
to generate a new deliverable, here
comes the fun part. Magic pen. Let me
show you. I want to select the previous
poster that we just made and then this
picture that I might want to use on the
next app and then pull my cursor here.
Click magic pen and here comes the
canvas. So basically it allows you to
doodle whatever you want just like
drawing and scripting on a real canvas.
Say I want to replace the placeholder
image. Then I will simply click pen and
then draw a line over it and then type
replace the placeholder image in the
poster with this image.
Then I will click the generate button to
send the whole canvas.
And wow, here it is.
In cues, every file you make becomes
contact for the next app. The more you
use it, the smarter it gets. From data
to design, from insight to lunch,
everything happens just in one place.
Try yourself and view your Q's workspace
by today.
Loading video analysis...