Re-envisioning what a day in the life of a software engineer looks like going forward in this agentic vision…
We’ve talked about why Ambient Code matters and what it means for your career. So what does your day look like when you’re shepherding code instead of writing it?
The Morning Dashboard (8:00 AM)
You grab coffee. You log into your Ambient Code Platform. You have a custom dashboard/cockpit view that shows (conceptually):
- GitHub notifications from overnight
- Agentic activity waiting for review
- Work blocked on your decisions
- Pull requests ready for merge
You’re not staring at an empty editor wondering where to start. You’re reviewing what your agents accomplished while you slept. There’s a button to unblock Claude after you’ve reviewed its work. You review, and work continues.
This is the first waterfall of your day.
Setting Up Your Sessions (9:00 AM)
You know what needs to happen today. You open five sessions. Five is my limit where my brain can track context without thrashing.
Each session gets a clear objective. You’re writing specs, not code. You’re defining what success looks like. You’re providing context the agent can’t infer on its own.
By 9:30, all five sessions are running. You’ve teed up async work. This is one of the cornerstone attributes of the ACP workflow. The agents are building. You’re not blocked waiting for compilation or test runs.
The Midday Check (1:00 PM)
Lunch happened. Now you’re back at the dashboard. What’s ready for review? What’s blocked? Where can you unblock progress with a clarification or a decision?
You spend 30 minutes moving things forward. Some work is ready to merge. Some needs another iteration. One session hit an architectural question you need to answer. You review, answer, and the agent continues.
The Afternoon Setup (4:00 PM)
You’re planning for the overnight run. What can bake while you’re not here? What needs long test cycles? What’s complex enough to benefit from multiple attempts? How many design samples should we create?
You configure some overnight sessions. You monitor for a bit to make sure they’re headed in the right direction. You submitted something from your ideas branch, too. Let’s see. You don’t babysit. You shepherd.
And this is your second waterfall of the day.
The Math That Matters
Two waterfalls per day. On average. Some days you get three. Some days external dependencies mean you only get one. The point is: work happens in parallel. Work happens overnight. Work happens when you’re in meetings.
The constraint isn’t typing speed – it’s decision-making bandwidth.
You can manage five concurrent contexts. That’s fine. The agents can manage 50 files across those five contexts. That’s the amplification.
What We’re Actually Measuring
The obvious metric is coding time saved. Dozens of percent. Maybe more. The honest answer is many (most?) teams don’t have a baseline yet. When work happens overnight, how do you measure the counterfactual?
Most people miss that the value isn’t just in code.
Agentic workflows drive business value from the stuff you’ve been putting off. The documentation nobody writes. The tests nobody has time for. The refactoring that’s been on the backlog for six months. The proof-of-concept that needs a week you don’t have.
I’ve heard of startups that are building this way. Others are coming. The question isn’t whether this works, its how fast you can adapt.
The Automated Future
In this current state, you’d running two waterfalls a day. You’re the orchestrator. You’re making the decisions and approvals.
Over time, patterns emerge. Decisions become policies. Policies become automation. Waterfalls start just happening.
That’s when it gets interesting: you’re not managing five sessions. You’re managing 50. You’re not checking in twice a day. You’re reviewing weekly. You’re not shepherding code. You’re shepherding systems.
The Real Shift
Having time to think. What a concept. Part of the benefits of agentic development is just that – we need to get out from under the systems we’ve ourselves built. I see this as a true opportunity to refactor our worklives to align with these new capabilities.
So you spend your morning reviewing instead of scaffolding boilerplate. You spend your afternoon on architecture instead of syntax. You spend your mental energy on problems that matter instead of problems that compile.
The code still needs a human in the loop. You’re that human. But you’re operating at a different level.
Welcome to the new day job!
Leave a Reply