Replicas runs coding agents in the cloud for engineering teams. A developer can ping the agent from Slack, assign it a Linear ticket, or tag it on a GitHub pull request. Replicas then starts an agent that clones the repository, installs dependencies, runs the real stack, and returns with a pull request.
The premise sounds simple: give an agent a sandbox and let it write code. In practice, that agent needs to run builds, boot databases, start development servers, and use the same tools an engineer runs locally. The environment underneath it cannot be only a place to execute code. It has to behave like a full computer.
The challenge
Agents need a complete development environment
Customers expect the agent to work like an engineer: check out a branch, run the build, start services, open the application, and verify that a change works before submitting it. That workload needs memory, disk, nested services, and an environment that stays available for the duration of the task.
“Agents need the same environment engineers have - a full computer. Those are big: a lot of memory, a lot of disk. Spinning up, running, and winding down a lot of large machines reliably matters a lot in use cases like ours.”
- Connor Loi, Founder, Replicas
Docker-based sandboxes slowed down the workload
Before E2B, Replicas used a Docker-based sandbox. That became a bottleneck because a common Replicas workflow is to run Docker workloads inside the agent environment. Nesting those services on top of an already containerized sandbox made customer workloads slower and heavier.
Reliability became urgent during fundraising
The sandbox reliability problem peaked near the end of Replicas' Y Combinator batch, just as the team was preparing to fundraise. Building the infrastructure internally was not realistic. Replicas evaluated raw AWS EC2 instances and five sandbox providers; E2B was the fastest and most reliable option they tested.
“I originally started with a competitor, but the search for something stable kept dragging on right as the team was heading into the end of its YC batch and preparing to fundraise. The sandbox infrastructure was basically the last thing I wanted to think about.”
- Connor Loi, Founder, Replicas
The solution
Warm environments turn a ticket into a running agent
Replicas captures each customer's configured workspace as an E2B Snapshot, including the cloned repository, installed dependencies, and development setup. When somebody pings Replicas in Slack or assigns it a Linear ticket, Replicas starts a new sandbox from that snapshot. The agent inherits the prepared environment and can begin work without rebuilding the customer's setup.
MicroVM isolation with room to grow
Replicas compared providers on more than feature checklists. It needed infrastructure the team could trust years into the future. E2B's MicroVM isolation, snapshots, open-source architecture, and responsive support gave Replicas a stable foundation for long-running coding agents.
“We wanted an offering that balanced a focus on open source with stability and fast support, and that was E2B.”
- Connor Loi, Founder, Replicas
It worked on the first try
The difference appeared immediately after the switch. E2B was the only provider that worked on the first try without additional flakiness to diagnose. Resolving the sandbox problem during the most important week of the YC batch let the team focus on its product and customers instead of infrastructure.
The results
- More than 60% lower customer latency for heavy Replicas workloads after moving to E2B.
- Three weeks saved on sandbox debugging during the final stretch of YC and fundraising.
- More than 20 YC companies now run engineering work through Replicas.
“Reliability is the only thing that matters, and lack of reliability was the main reason for churn at Replicas. You want to pick infrastructure that's never going to let you down, so you have time to work on the thing that matters most: the agent experience.”
- Connor Loi, Founder, Replicas
Looking ahead
Replicas is building toward fleets of agents that run engineering workflows end to end: triaging tickets, writing code, responding to CI failures and review comments, and merging changes without anyone in the room. As Replicas serves larger engineering teams and enterprise contracts, it is leaning further into E2B's MicroVM isolation and flexible deployment options so every agent receives an environment engineers can trust.