In the second part of a two-part series, Jon Block, Founder & Principal Consultant, Syllepsis delves further into why every team in your organisation should be using AI coding agents.
In Part 1, we explored how AI coding agents could become the universal operating system for knowledge work. We discussed how finance teams and marketing departments might use these platforms to build systematic, repeatable automation through structured English instructions.
But understanding a possible future is one thing. Actually making the shift is another.
What This Means for You
If you’re looking to get ahead in your career – regardless of discipline – I’d recommend you start experimenting with an AI coding agent as your operating system right now.
Not for your company’s work yet. Start with something personal. Maybe you’re job searching – use Cursor or Windsurf to manage research, track applications, generate tailored cover letters and organise interview preparation. You’ll learn how these platforms work, discover where AI can systematically accelerate your work, develop intuition for good prompts and build skills that will be essential in the companies moving fastest.
When you land that next role, you’ll be the person who knows how to bring this approach to your new team. You’ll be the one who can show finance, marketing or operations how to engineer their workflows using Software 3.0 principles. That’s going to be an extraordinarily valuable skillset.
Why This Feels Harder Than It Should
“This sounds incredibly technical. I’m not an engineer. Why would I use something called an Integrated Development Environment?”
Fair reaction. And honestly, the terminology doesn’t help.
But here’s what’s actually happening: we’re in an awkward transition period. The most powerful tools for working with AI happen to be the same platforms engineers use for writing code. Not because non-engineers should be using engineering tools, but because these platforms got there first. They already had the infrastructure that turns out to be exactly what you need for systematic AI automation.
Think about the early days of the internet. To send an email in 1990, you needed to understand command lines, configure SMTP servers and memorise cryptic addresses. Was email inherently technical? No. But the tools were, because engineers had built them for engineers. Then Outlook arrived. Then Gmail. The underlying technology didn’t change but the interface evolved to match how normal people actually work.
That’s where we are now. The capabilities are here but they’re wrapped in interfaces designed for a different audience. New tools are emerging – platforms specifically built for managing prompts, context and AI workflows without assuming you’re a software developer. The infrastructure will remain. But the interfaces will evolve to feel natural for finance teams, marketing departments and operations leads.
Learning to use an IDE feels intimidating right now. But the underlying skills – writing clear instructions, organising information systematically, testing that things work and collaborating with colleagues – those are skills you already have. You’re just applying them in an unfamiliar interface.
And that interface is already getting better.
What This Means for Companies
If you’re leading a business – particularly if you’re still early-stage or in high-growth mode – this is your moment. Unwieldy organisations will struggle with this transition. They have entrenched processes, risk-averse cultures and procurement hurdles. The idea that your finance team should be working in an IDE will sound insane to most large company leadership.
Which means you can move faster.
You can become the company or team where everyone learns to use Context Engineering and Prompt Engineering as core skills. Where systematic automation isn’t a technical speciality but how work gets done. Where your 20-person team operates with the output velocity of a 200-person traditional organisation.
This requires investment in training and accepting there will be a learning curve. But most critically, it requires thoughtful change management – helping teams understand why this matters, supporting them through the transition and building the organisational muscle memory that makes this way of working feel natural rather than forced. The companies that make this shift now, with deliberate strategy rather than chaotic adoption, will build competitive advantages that compound over time.
The Uncomfortable Truth
Look, I understand this sounds foreign. I understand that telling your account management team they need to learn version control and write prompts in markdown files feels like a bridge too far.
But here’s the uncomfortable truth: this is happening whether organisations embrace it deliberately or stumble into it reactively.
The question isn’t whether knowledge workers will need to become fluent in Context Engineering and Prompt Engineering. They will. The only question is whether your organisation helps them get there systematically or whether they do it chaotically as a survival response when competitors start moving 10x faster.
The Three Waves Ahead
In Part 1, I said this is as fundamental as the transition from memos to email. But unlike email, AI is still evolving rapidly.
I believe we’re living through the first of three major waves:
Wave 1: Human-AI Hybrid (Now-2030) – AI is a powerful collaborator but humans remain essential for oversight, quality control and strategic direction. The organisations winning right now are using AI coding agents as we’ve described – systematising workflows, building reusable prompt libraries and treating knowledge as engineered artefacts.
Wave 2: AI-Led Coordination (2030-2040) – AI might not just execute instructions but coordinate entire workflows autonomously. Requirements could be defined, interpreted and executed by AI systems working together. The human role might shift from execution oversight to strategic stewardship. Traditional software engineering roles could transform fundamentally.
Wave 3: Post-Engineering Cognition (2040+) – Eventually, the distinction between “building applications” and “being the application” might dissolve entirely. AI systems could become the cognitive layer that mediates all interaction between people, systems and data.
I won’t pretend to know exactly what Waves 2 and 3 look like. Nobody does. The future remains uncertain and there are huge questions about what this means for employment, skills and society that I can’t answer.
But here’s what I suspect: if you’re not on board with Wave 1, you won’t survive Wave 2.
The organisations that master this first transition – that learn to treat English instructions as engineered artefacts, that build systematic automation capabilities across all departments – those organisations will be positioned to adapt when Wave 2 arrives.
The ones still dismissing this as “just for engineers”? They won’t make it.
This isn’t about job security in some distant future. This is about building the capabilities now that position you to ride the waves that are coming. Each wave builds on the last. You can’t skip ahead.
The Choice
The shift is universal. The advantage goes to those who move first.
We’re at the beginning of a transformation as fundamental as the one that happened when email replaced memos. The organisations that recognise the moment they’re in – that invest in helping their people adapt systematically rather than waiting until change is forced upon them – those are the organisations that will define productivity for the next decade.
The question isn’t whether this is coming. It’s already here.
The question is: are you one of the organisations getting there first?
