My Complete Agentic Development and Productivity Setup in 2026

I’ve spent months working in a way that’s nothing like what I was doing a year ago. I don’t open an IDE with the code in front of me waiting for it to compile. I don’t review changes line by line. My workflow is completely different.

I use AI agents as my main work tool. And not just for programming.

This is the setup I use every day, how I combine it, and why it works.

Disclaimer: My work doesn’t follow a typical company workflow, so I don’t recommend copying it exactly—I’m sharing it in case it inspires you. I try to push tools to their limits to see what works and what doesn’t. If you want to apply AI to professional software development, I recommend you check out AI Expert.

My Base of Operations: Codex App

All my development goes through Codex App. It’s where I have all the projects I’m working on open: my website, my tools, my automations.

The model I use is GPT 5.3 Codex High, the latest available. It’s my favorite for code.

What has changed my workflow the most is that I can launch tasks in parallel. While Codex is working on a refactor for one project, I’m reviewing the PR of another. Or launching a skill that publishes a video on YouTube. Or preparing the next deploy.

It’s not the classic flow of “write code → run → fix → repeat.” It’s more like managing a team: I decide what needs to be done, give context, launch the task, and review the result. Sometimes in several projects at once.

I even have some projects being created automatically thanks to Codex automations.

Claude Code: Design and Chatbot

Claude Code is my perfect complement for what Codex doesn’t do so well: user interfaces.

Codex is amazing at generating logic, but when it comes to designing a UI, it falls short. My usual flow: I let Codex generate the first functional version of an interface, and then I open Claude Code to polish it. Sometimes I use Pencil and its MCP alongside.

The other big use for Claude Code is my Telegram chatbot. It’s my personal assistant: I talk to it by text or voice and it manages my calendar, tasks, invoices, emails (always as drafts, it never sends anything on its own), content, and a ton of other things.

I alternate between Claude and Codex models for the chatbot, because I’d like to pay for just one subscription, but I like Claude’s results a bit more for this kind of conversational interaction.

Agent Skills: The Game Changer

Skills have been the biggest change. I won’t go into detail because I already wrote about them in depth, but in short: they’re Markdown files that tell the agent exactly how to behave for a specific task.

I have skills for all kinds of workflows: posting on social media, managing invoices, processing emails, generating newsletters, publishing videos on YouTube, managing tasks… Most have nothing to do with programming, though some do. I encourage you to check out the original post.

What used to be 10 manual steps (open tool, copy data, format, send…) is now a single command. And they work in Claude Code, Cursor, and Codex.

Crons: My Team Working While I’m Away

This is what really makes a difference in productivity, and above all helps me avoid procrastination. I have agents that run automatically at specific times. Here are some examples:

Time	Automation	What it does
9:00	Content curation	Reviews news, trends, and updates in AI for devs. Prepares a daily digest for me.
9:00	Morning sweep	Checks emails, pending tasks, Slack, and WhatsApp. Tells me what needs my attention.
9:15	Content proposal	Depending on the day of the week, suggests the corresponding post, article, or video with a pitch and a draft.
10:00	Issue triage	Reviews open issues in my automated repo, prioritizes them automatically, and gives me an activity summary.
15:00	Afternoon sweep	Same as the morning: reviews what came in during the day.
Friday 16:00	Weekly newsletter	Gathers everything published during the week and generates a newsletter draft ready for review.
1st of the month	Monthly report	Cross-references content published on X with website traffic and sales to see what really works.
Monday 9:00	Weekly strategy	Analyzes the performance of my posts from the last 15 days and suggests new ideas.

These aren’t simple scripts. They’re full agents with context, memory, and decision-making ability. If there’s nothing pending, they don’t bother me. I have a Telegram group with different topics where each thing gets sent.

I tweak this setup often, so they help me without creating too much noise. And it’s helped me walk more and spend less time sitting at the computer. I solve a lot of things from my phone 😄

And at night, while I sleep, another agent explores ideas, does experimental mini-projects, and leaves me a log of what it built. Some mornings I’ve woken up to a functional tool that didn’t exist the night before.

What I’ve Learned

Context is everything. An agent without context creates noise. An agent with your configuration, your skills, and your memory creates real value. Investing time in setting up your environment pays off from day one.

Automate the repetitive, not the creative. Crons are for predictable tasks. I make the important decisions, with the information the agent prepares for me. The AI doesn’t decide what content I publish or what architecture I use. It gives me options and I choose, suggest alternatives, and adapt.

Fewer tools, better configured. I’ve tried dozens. I stick with Codex + Claude Code + well-integrated skills + crons. I don’t use subagents, hooks, or other fancy tools. For now, I don’t see the point.

The role changes, it doesn’t disappear. I write less code line by line, but design more systems. I do fewer manual tasks, but make more decisions. And the responsibility for the outcome is still mine.

What’s your current setup? What tools do you use in your day-to-day for development and productivity?