I Built a Tool to Manage Multiple AI Agents at Once

The DevOps Toolkit (Viktor Farcic)
The DevOps Toolkit (Viktor Farcic)May 4, 2026

Why It Matters

Orchestrating multiple AI agents boosts development throughput while turning engineers into managers, demanding new tooling and reshaping team dynamics.

Key Takeaways

  • Managing multiple AI agents transforms engineers into de facto project managers.
  • Separate git work trees per agent prevent file conflicts on local machines.
  • Tool must be terminal‑based, keyboard‑driven, and integrate with existing agents.
  • Agent Deck provides a dashboard, pane creation shortcuts, and real‑time status.
  • Visibility and quick switching are essential to avoid chaos in parallel execution.

Summary

The video introduces a new workflow where software engineers shift from coding to managing a fleet of AI agents. By running several agents in parallel, developers assume roles traditionally held by project managers, tech leads, and product owners, supervising, directing, and validating work that the agents generate.

To make this model viable, the presenter outlines three core activities—planning, supervision, and validation—and derives concrete tool requirements. Agents must operate in isolated git work trees to avoid file clashes, the orchestration interface should live inside the developer’s preferred terminal, and it must respect existing agent clients rather than replace them. A lightweight, keyboard‑centric dashboard that surfaces each agent’s status, recent prompts, and tool calls is essential for maintaining oversight without constant context‑switching.

The speaker demonstrates "Agent Deck," a terminal‑only application that fulfills those criteria. Using shortcuts like Ctrl‑N to spawn panes and Ctrl‑D to enter command mode, users can spin up multiple agents, assign them distinct branches, and monitor their progress on a single screen. The UI compresses agent cards as the roster grows, ensuring the whole team remains visible at a glance, while still allowing deep‑dive views with automatic log and file monitoring.

The shift toward AI‑agent orchestration redefines engineering work, pushing many developers into managerial functions and prompting organizations to rethink team structures. Tools like Agent Deck enable scalable parallel development, but they also raise questions about the future of traditional management roles and the balance between human oversight and autonomous AI execution.

Original Description

Running multiple AI agents in parallel isn't just a productivity upgrade — it's a fundamental shift in how we work. When you go from one agent to many, you stop being a software engineer and start becoming a project manager, tech lead, and architect all at once. This video breaks down what that transition actually looks like, what a tool needs to support it properly, and introduces Agent Deck, a purpose-built TUI that makes orchestrating multiple agents practical without forcing you to abandon the terminal or agent clients you already love.
The hands-on demo walks through spinning up six agents working simultaneously on separate features, each in its own git worktree, with a live dashboard showing every agent's status, recent activity, and whether it needs your attention. From keybinding-driven navigation to automatically surfacing relevant context when zooming in on a specific agent, Agent Deck is designed around the real demands of managing an AI team — staying out of the way when things are running smoothly and making it instantly obvious when something needs you.
#AIAgents #TerminalTools #AgentDeck
Consider joining the channel: https://www.youtube.com/c/devopstoolkit/join
▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬
▬▬▬▬▬▬ 💰 Sponsorships 💰 ▬▬▬▬▬▬
If you are interested in sponsoring this channel, please visit https://devopstoolkit.live/sponsor for more information. Alternatively, feel free to contact me over Twitter or LinkedIn (see below).
▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬
▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬
▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬
00:00 Running Multiple Agents
00:45 Managing AI Agents in Parallel
05:07 Multi-Agent Tool Requirements
12:30 Agent Deck Hands-On Demo
20:05 AI Agents: What Comes Next

Comments

Want to join the conversation?

Loading comments...