Automate the Edges First
Key Points
- Focus on automating the “edges” of a workflow—data preparation, QA, synthesis, and handoffs—because AI can cut cycle times by 70‑90% there, delivering the biggest immediate ROI.
- Core processes are often riddled with ambiguity, exceptions, and tribal knowledge, so trying to automate them first leads to stalled agents, scope creep, and frustrated teams.
- Treat edge automation as a low‑risk entry point: evaluate how you currently collect, clean, and normalize context, and let LLMs handle those repetitive steps before tackling the full workflow.
- Leverage LLMs for quality checks and summarization tasks such as consolidating ticket discussions, templating outputs, or grouping relevant information—high‑value work that’s simple for AI but time‑intensive for humans.
- Once edge automation proves effective, use the streamlined outputs to package deliverables (briefs, reports, etc.), creating a repeatable pipeline that can later support more ambitious core‑workflow automation.
Sections
- Automate the Edges First - The speaker urges teams to focus AI automation on peripheral tasks such as data preparation, quality assurance, synthesis, and handoffs—where 70‑90% cycle reductions are possible—rather than tackling the core, ambiguous workflow, thereby avoiding stalled agents, bloated scope, and frustrated stakeholders.
- Automating High‑Friction Workflow Edges - The speaker argues that AI agents should initially focus on the coordination‑heavy, high‑friction edges of a process—tasks that are low‑judgment, data‑ready, and easily recoverable—because improving these spots delivers immediate value without disrupting the core workflow.
- Automating Workflow Edges for Trust - The speaker suggests that real AI transformation comes from first automating peripheral tasks like intake, data pulling, QA checklists, and synthesis to build reliability and trust before attempting to replace core processes.
Full Transcript
# Automate the Edges First **Source:** [https://www.youtube.com/watch?v=B3rSU7XROrg](https://www.youtube.com/watch?v=B3rSU7XROrg) **Duration:** 00:08:05 ## Summary - Focus on automating the “edges” of a workflow—data preparation, QA, synthesis, and handoffs—because AI can cut cycle times by 70‑90% there, delivering the biggest immediate ROI. - Core processes are often riddled with ambiguity, exceptions, and tribal knowledge, so trying to automate them first leads to stalled agents, scope creep, and frustrated teams. - Treat edge automation as a low‑risk entry point: evaluate how you currently collect, clean, and normalize context, and let LLMs handle those repetitive steps before tackling the full workflow. - Leverage LLMs for quality checks and summarization tasks such as consolidating ticket discussions, templating outputs, or grouping relevant information—high‑value work that’s simple for AI but time‑intensive for humans. - Once edge automation proves effective, use the streamlined outputs to package deliverables (briefs, reports, etc.), creating a repeatable pipeline that can later support more ambitious core‑workflow automation. ## Sections - [00:00:00](https://www.youtube.com/watch?v=B3rSU7XROrg&t=0s) **Automate the Edges First** - The speaker urges teams to focus AI automation on peripheral tasks such as data preparation, quality assurance, synthesis, and handoffs—where 70‑90% cycle reductions are possible—rather than tackling the core, ambiguous workflow, thereby avoiding stalled agents, bloated scope, and frustrated stakeholders. - [00:03:13](https://www.youtube.com/watch?v=B3rSU7XROrg&t=193s) **Automating High‑Friction Workflow Edges** - The speaker argues that AI agents should initially focus on the coordination‑heavy, high‑friction edges of a process—tasks that are low‑judgment, data‑ready, and easily recoverable—because improving these spots delivers immediate value without disrupting the core workflow. - [00:06:35](https://www.youtube.com/watch?v=B3rSU7XROrg&t=395s) **Automating Workflow Edges for Trust** - The speaker suggests that real AI transformation comes from first automating peripheral tasks like intake, data pulling, QA checklists, and synthesis to build reliability and trust before attempting to replace core processes. ## Full Transcript
I want to let you in on a little secret
around AI automation and agents.
Automate the edges first. And I'll get
into what I mean there. Most teams burn
months trying to automate the core of
their work, the thing the humans already
do pretty well. The real leverage often
comes from automating the edges, the
data preparation, the QA, the synthesis,
the handoffs. AI can quietly compress
cycles here by 70 80 90% but most people
don't start here. I want to note that
this is different from the problem space
you pick. So if you're saying, Nate, I
thought you tell us to pick something
important to work on, 100% I do. I think
you need to pick things that matter for
AI. I'm saying once you do, think about
the edges of the work because there's
tons of leverage around that valuable
problem space in the edges of the work.
And so I get the automate everything
vision, especially if you have a core
workflow. But keep in mind that most
core workflows start out when you face
them containing ambiguity. They contain
exceptions. They contain tribal
knowledge. Teams underestimate the
hidden state and tend to overestimate
model reliability, especially if you
haven't built an AI agent automation
before. What does this lead to? It leads
to stalled agents. It leads to bloated
scope. It leads to frustrated
leadership. frustrated engineers,
endless QA. If you are trying to
automate the core first, it's kind of
like trying to build a self-driving car
before you've invented cruise control.
My challenge for you, when you pick a
valuable workflow to automate, if this
is your first AI agent job, figure out
the edges of your workflow and just
test, just see if there is something
here that gives you a lot of bang for
your buck. Look at data preparation. How
do you collect context for this workflow
today? How do you clean your data
inputs? How do you normalize your
formats today? Is that a manual process
before you even get into the core
workflow? Look at QA. How are you
checking for dness, completeness,
quality, consistency, obvious errors?
Something that an LLM as judge can
perhaps easily do that doesn't require
doing the whole workflow. Synthesis is
another great example. Let's say that
all you're trying to do is not automate
the full workflow, but you're picking a
valuable part of it and you're saying, I
just need to summarize information to
date. I want to summarize the discussion
thread in the JUR ticket and update the
description, right? I want to summarize
and synthesize information that is
relevant in the workflow and communicate
it over here. That can also look like
grouping information. It can also look
like templating output. So, you have the
information and you're just writing it
to template. Super valuable work. often
takes a lot of human time but not super
hard for the LLM and is a valuable edge
to go after. Another edge to go after
the packaging of the work. How do you
convert the work into deliverables? It's
done. How do you get into a brief? How
do you get into a report? Especially now
with the advent of Nano Banana, with the
advent of Gemini 3, with Opus 4.5,
working on PowerPoint skills for longer,
harder, you have options to get all the
way to finish deliverable that you did
not have 3 months ago. That is another
edge that you can start to look at.
Coordination is another edge that often
has a ton of value, especially in tribal
knowledge situations. coordination often
resides in someone manually pulling
information here, talking to someone,
then putting it over here. If you can
pick up that piece where you have the
information and you just need to get it
over to point point B from point A, that
is often very very very valuable. So why
do I suggest edges of the work? They're
high friction because typically workflow
is least frictional at the core and most
frictional at the edges. That's just a
general observation anyone will tell you
having done workflows. It's the edges
that are often the worst. It's also
often a low judgment task because all of
the inputs are ready, which is perfect
if you're just starting out on AI
agents. And that means they are perfect
for LLMs. Even when LLMs are imperfect,
you should not assume that your first AI
agent is perfect. You should assume it's
imperfect and it needs to deliver value
anyway. This also means that errors are
often recoverable and cheap because the
humans doing the core of the workflow
were doing those edges before. And if
there's an exception that occurs, they
can pick that up easily. You have the
chance to look at the data. You have the
chance to fix it. And you have the
chance to come back and make your agent
better. This also means, by the way,
that you are not abandoning the core of
the workflow. If your goal ultimately is
to have an AI agent sit at the heart of
the workflow, you get a clean path into
that by attacking the edges. If you own
QA, if you own handoff, if you own data
inputs and data preparation, you are
well positioned to have the knowledge
you need to do the AI agent at the heart
of the workflow, that may be your
ultimate goal. You position yourself by
being at the edges of a core valuable
workflow. You position yourself to
attack the heart of that workflow next
and then to snowball those gains across
the org. Because really, what you're
doing is twofold. You're not just going
after this core workflow. You are
teaching yourself and teaching the org
how AI automation ought to work. And
this is the part that almost nobody says
out loud. You are not just doing a
technical project. You are doing an
upskilling project. Not just for the
engineers building the agents, but for
the humans involved. And the humans
involved tend to have a lot of tribal
knowledge. They tend to be fingertippy
on the work. If it's a valuable
workflow, they need to be able to be
confident that your AI automation task
with the agent will not cost them that
fingertippy feeling on the work. They
are crafts people. Make sure they know
where their craft can be practiced. If
the part of the work that is highly
valuable in this workflow is the highle
understanding of the customer history
over multiple years and how you nuance a
particular response to the customer.
That's a customer success example. You
want to automate around that so that the
customer service agent can apply that
knowledge efficiently with their full
intuition with their full human memory
of the relationship and not be
distracted by other stuff. And so when
you start by attacking the edges, you
are reminding the people doing the work
that their fingertippy feeling for the
work is valuable, that they are worth
having involvement in the work because
of the craft they bring. That is
critical because if you lose that trust,
they will not be inclined to share with
you all of the secrets of the art that
you need for the rest of the workflow.
You need to look at AI agent building as
an exercise in trust. There is no
substitute. And so I'm going to argue
that the real leverage hides outside the
core. It hides in stuff like intake, in
data pull, in QA checklists, in
synthesis, in packaging. You get the
idea. And when you do this, reliability
can go up. You have less risk. You're
attacking a core workflow. You're
showing gains and you're earning the
trust of everyone involved to get where
you want to go. This leads to teams
winning fast. So if you want to apply
this tomorrow, pick a workflow that you
touch every single week that's valuable.
Map the edges. Where do you waste time
prepping? Where do you check for errors?
Where do you hand off repeatedly? Where
do you summarize over and over? Pick the
simplest edge. get into Chad GBT to
claude to Gemini and focus on thinking
about how you build a simple solution.
It's okay if it's semmanual to start and
you start to automate from there. That's
fine. The point is that you're
approaching it correctly and then you
can build the automation edge inward.
Automation does not start with replacing
the core unless you have a very
experienced engineering team. It starts
with reclaiming the edges. So if you
automate three or four edges in a row
and you're starting to feel good, you
don't need the full grand vision. The
workflow itself will reveal the answer
to the correct place of automation and
the correct place of human expertise.
And that's how real AI transformation
happens. And I wish we talked about it
more. You tell me where are you looking
to automate