Working environments for Claude Code, Codex, and Cursor. Connect a repository →

Give twenty engineers the execution capacity of two hundred.

NullDeref gives your coding agents a working replica of each repository’s real target. They build, debug, test, and benchmark there autonomously—then return reviewable work.

Runs inside your controls PRs remain the review gate No per-seat tax

A working environment for every target

Dockerservices and toolchains
Linux / kerneltrees, modules, drivers
VM / hypervisorbootable system replicas
Embeddedfirmware and target boards
Onboard the repository from GitHub or GitLab. One connection brings in the code, permissions, and pull-request workflow. Add Jira to turn tracked work into agent runs and return results to the same ticket.

Keep your coding agent. Fix what it runs against.

Claude Code, Codex, and Cursor already know how to write code. NullDeref supplies the part they were missing: a dependable environment where the code can compile, boot, fail, recover, and prove the result.

Claude Code Anthropic
Codex OpenAI
Cursor SpaceX

Every task closes the loop before it reaches review.

The agent changes the code, compiles it, diagnoses the break, runs the test suite, and measures the result. Winners become pull requests. Regressions are reverted. Engineers review evidence instead of supervising terminal sessions.

01 build
02 debug
03 test
04 benchmark
InputA Jira ticket or engineering goal
GateCorrectness, tests, and target benchmark
OutputA reviewable PR with the run attached

Six controls for autonomous engineering work.

Start with one repository. Add the operating layer your team needs as agent work scales across the organization.

10.9× faster after 16 autonomous loops.

The agent was given a straightforward goal: speed up an image-processing task without changing its output. It ran the code, measured each attempt, kept improvements, and discarded regressions until the performance gate passed.

10.9×
faster than the starting version
16
build–measure loops to reach the result

Source

Pay per repo. Not per seat.

Copilot, Cursor, and Claude charge the human. NullDeref charges the repository—the durable unit that needs an environment, policy, and execution history. Every engineer on a connected repo is included.

Volume

Org

Contact

Twenty-five or more repos. One picture of the org, not a stack of invoices.

  • Everything in Repo
  • Scope — agent readiness across the org
  • Meter — token economics by successful task
  • Pulse — evidence-backed answers across engineering
  • Volume pricing
Contact us

Custom

Enterprise

Contact

SSO, Jira, isolated Red labs, and terms that survive a security review. The floor is a conversation.

  • Everything in Org
  • Red — autonomous campaigns with exploit proof
  • Jira issues back from the loop
  • SSO, VPC, custom terms
Contact us

Published list price, not a negotiated floor. Agent usage from Anthropic, OpenAI, or Cursor is billed by those vendors. Enterprise controls, private networking, and custom terms are scoped separately.

Put an agent to work in the environment your code actually needs.

Tell us where you want to start. We’ll confirm the target environment, repository controls, and the first acceptance gate with your team.

1
Connect GitHub, GitLab, or your Jira workflow.
2
Define how the repository builds, runs, and proves correctness.
3
Send completed work through your existing PR and CI process.

Choose where to start.

Use a work email so we can discuss the repository and its target environment.

Which plan?

Which product first?

Something went wrong sending that — try again, or email us directly.

no mailing list · direct follow-up from our team