NewWhat CI Actually Looks Like at a 100-Person Team

The AI DevOps Engineer.

Automating the DevOps work your engineers shouldn't be doing manually anymore. Supply chain security, flaky CI, slow builds, and anything else specific to your stack.

Trusted by engineering teams shipping 10,000+ PRs per week

PostHogMetabaseLuminaiInngestClipboard HealthBlaxel
From the team behindDocker&Dagger
YBacked by Y Combinator

Three agents always on.Custom automations for everything else.

Security, Reliability, and Performance ship built-in.
Add custom automations for any DevOps work specific to your stack.

Always on

Security

Reviews dependency PRs. Pins safe versions, surfaces only the CVEs actually exploitable in your code.

Always on

Reliability

Diagnoses CI failures, fixes flaky tests, ships fix PRs.

Always on

Performance

Cuts build time: caching, parallelism, slow tests to drop.

Custom

Datadog alert

Pulls traces, logs, recent deploys → posts likely cause.

Custom

Sentry triage

Groups errors by release, finds suspect commit, drafts a patch.

Custom

Cloud deploy

Checks runtime logs, IAM, config drift → repair plan or rollback.

Custom

Migration safety

Reviews PRs that touch migrations. Flags blocking ops, table locks, or downtime risk before merge.

Custom

Anything else

Trigger from CI/CD, infra, Slack, Linear, webhook, or a schedule.

Trusted in production.

Anyshift

Mendral saved our a** during the Trivy supply chain attack. The agent had pinned the secure version two weeks before, so we avoided rotating every key across our infra.

Stephane Jourdan
Co-founder & CTO
PostHog
104
fixes from Mendral accepted in a month

Mendral has been super impressive for us so far!

Tim Glaser
Co-founder & Co-CEO
Read the full story
Metabase
240K
CI runs per week

Mendral has been super helpful for us. It keeps our CI efficient and manageable at scale.

Vamsi Peri
VP of Engineering

Signal in. PR or evidence out.

Mendral plugs into the tools where engineering work starts, keeps the investigation tied to the original signal, and leaves a trail your team can audit.

Event01

A signal arrives

A check fails, an alert fires, a deploy breaks, a dependency changes, or a schedule runs.

Investigation02

Mendral pulls the right context

It reads logs, traces, commits, cloud state, previous fixes, and repo conventions.

Output03

Your team gets a PR or evidence

Mendral returns the PR, review, or handoff your team expects. If code shouldn't change, it says exactly what to do next.

Feedback04

Every review trains the next run

Accepted fixes, rejected approaches, comments, and playbooks shape the next investigation.

SIGNALS INACTIONS OUTFEEDBACK IN▸ MEMORYknowledge graphREACTinvestigateREMEDIATEfix & shipREMEMBERaccept · reject · refinegithubcisentrydatadoggcpschedulesfailuresperformancesecurityflakypull_requestsslackdashboardreviewscommentsplaybooks
Time to Green

red → green, faster

Change Failure Rate

fewer regressions shipped

Ship Velocity

deploys per week, up

73% of CI failures auto-matched to known issues across millions of jobs.

From the team behind
Docker & Dagger

DockerDagger

We spent over a decade shaping DevOps and CI/CD. We built Docker from the ground up, then co-founded Dagger to modernize CI. Now we're reinventing software delivery for the AI era.

Sam Alba

Sam Alba

Andrea Luzzardi

Andrea Luzzardi

Get early access today.
First fix within minutes.

Drop your work email and we'll reply within a business day. Once Mendral is connected to your CI, the first fix typically lands within minutes of onboarding.