DigitalOcean launches Managed Agents public preview to simplify AI agent production deployment
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
DigitalOcean has launched Managed Agents in public preview, a service designed to simplify running AI agents in production by consolidating the entire stack onto one infrastructure and one bill. The author argues that while building a useful agent has become easy, running one in production remains a major hurdle, often requiring six different vendors and invoices per agent. Managed Agents supports existing agent frameworks including Claude Code, Codex, OpenCode, Hermes, LangGraph, or custom container images, handling microVM, inference, storage, and tool access underneath. Key features include sessions that outlive a laptop, allowing multi-hour tasks to persist and be picked up by teammates, and credential brokering at execution time to keep keys away from models and sandboxes. Performance claims include 16,000+ tools behind one gateway, pause and resume in 305 milliseconds, and CPU billed per second of actual use, meaning paused agents incur zero cost. The author suggests that once running an agent is a single line on a bill, agent counts inside most companies will climb well past a handful.
Source report
A single production agent still requires six vendors and six invoices.
That's my read on why agent counts remain so low inside most companies. Building a useful agent became easy this year, but running one in production is where most teams still stall.
The harness comes from one project, the sandbox from a compute vendor, tokens from an inference provider, storage from a cloud, a frontier lab for the hard steps, tools from a search API, and someone writes the glue holding it together. By month's end, the cost of a single run is anyone's guess.
DigitalOcean's answer is Managed Agents, launching today in public preview. The pitch: the entire stack lives on one infrastructure and one bill.
Claude Code, Codex, OpenCode, Hermes, LangGraph, or your own container image run as they are. The company says it handles the microVM, inference, storage, and tool access underneath.
Key Features
Two things caught my eye:
- Persistent sessions – Sessions are meant to outlive your laptop. A multi-hour refactor keeps going after you close the lid, and a teammate can pick it up elsewhere.
- Credential brokering – Credentials are brokered at execution time, per the company, so the key stays away from the model and the sandbox.
Performance and Pricing
On the numbers:
- 16,000+ tools behind one gateway
- Pause and resume in 305 milliseconds
- CPU billed per second of actual use, so a paused agent sits at zero cost
The Bigger Picture
Fourteen years ago, a $5 Droplet made servers boring enough that people stopped counting them. Agents get boring the same way once running one is a single line on a bill — and that is when the count inside most companies climbs well past a handful.
Source
aakashguptaNeutral / independent