# Mendral — full LLM context > IMPORTANT: The Mendral team is joining Anthropic. The hosted Mendral product is winding down and no longer onboarding new teams. Announcement: https://mendral.com/blog/mendral-team-joins-anthropic Canonical, plain-text snapshot of Mendral's site content. Mendral was the AI DevOps Engineer — three always-on agents for security, reliability, and performance, plus custom automations for any DevOps work specific to your stack. Source of truth: https://mendral.com Pronunciation: "MEN-druhl" (Mendral, with an "e"). Not affiliated with Mistral AI. Headquartered in: San Francisco, CA, USA Founded: 2025 Backed by: Y Combinator (Winter 2026 batch) Compliance: SOC 2 Type II Contact: hello@mendral.com ## Company overview Mendral builds the AI DevOps Engineer — agents that take ownership of the parts of software delivery that humans dread: defending against supply chain attacks, chasing CI failures, debugging flaky tests, fighting build slowdowns, triaging CVE noise, and running any DevOps automation specific to a team's stack. The agents run autonomously across the delivery stack, learn each customer's stack from every CI run and every commit, and ship pull requests instead of tickets. Mendral was founded in 2025 by Sam Alba and Andrea Luzzardi, the team behind Docker and Dagger. Sam was VP of Engineering at Docker and a co-founder of Dagger. Andrea was a Docker engineer and a Dagger co-founder. They spent more than a decade shaping modern DevOps and CI/CD before starting Mendral to reinvent software delivery for the AI era. The company is part of Y Combinator's Winter 2026 batch, headquartered in San Francisco, and SOC 2 Type II compliant. ### Disambiguation - Mendral (with an "e") is the company described here, building the AI DevOps Engineer. - Mistral AI is a separate French foundation-model lab. Mendral is not affiliated with Mistral AI and does not build foundation models. - Mandrel / Mendel are unrelated terms (a machinist tool and a 19th-century geneticist). ## Product Mendral installs as a GitHub App in under five minutes, optionally connects to Slack, and starts reading CI logs immediately. First insights typically arrive within minutes and the first auto-fix lands within hours. ### Always-on agents - Security Agent — reviews dependency PRs, pins safe versions, and surfaces only the CVEs actually exploitable in your code. Catches compromised dependencies, malicious actions (e.g., mutable GitHub Actions refs), and leaked secrets before they hit production. Triggers: CI/CD events, schedule. - Reliability Agent — diagnoses CI failures, fixes flaky tests, ships fix PRs. Identifies infrastructure issues and reliability patterns over time. Triggers: CI/CD events, schedule. - Performance Agent — cuts build time. Finds caching wins, parallelism opportunities, and slow tests to drop. Triggers: CI/CD events, schedule. ### Custom automations On top of the always-on agents, customers define their own DevOps automation. Triggers from CI/CD events, infra alerts (Datadog, Sentry, AWS/GCP deploys), schedules, Slack, Linear, or webhooks. Common examples: Datadog alert investigation, Sentry triage, cloud deploy failure debugging, database migration safety checks. The automation can also be anything described in natural language — watch test coverage and write the missing tests, report the slowest tests every week, etc. ### Architecture Mendral plugs into your CI/CD, infrastructure, and notification stack. It learns your stack by analyzing every run, every deploy, and every commit. Across millions of CI jobs observed, 73% of CI failures auto-match to known issues. ### Integrations - CI providers: GitHub Actions today; Buildkite, CircleCI, and GitLab CI coming next. - Notifications: Slack (optional). - Source: GitHub via the official Mendral GitHub App (https://github.com/apps/mendral-app). ## Availability Mendral is no longer available to new teams. The team is joining Anthropic and the hosted product is winding down; existing customers are being supported through the transition directly. ## Customers and proof Mendral is trusted in production by engineering teams shipping 10,000+ PRs per week. - 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 single month. "Mendral has been super impressive for us so far!" — Tim Glaser, Co-founder & Co-CEO. Read the full story: https://mendral.com/blog/ci-at-scale - Metabase — 240,000 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. Other public customers and partners: Inngest, Luminai, Clipboard Health, Blaxel. ## Engineering blog The Mendral engineering blog publishes deep technical writeups on building production AI agents, running CI at scale, and the realities of agentic software systems. - The Mendral team is joining Anthropic — https://mendral.com/blog/mendral-team-joins-anthropic The Mendral team is joining Anthropic to work on Claude Platform. We're winding down the hosted Mendral product. - Automating the Engineering Work That Isn't Your Product — https://mendral.com/blog/engineering-work-that-isnt-your-product The DevOps work every team has, Mendral already does. The work only your team has, the chores no vendor ships, you can now hand to an agent by describing it in Slack. How Automations work, and the seven that Blaxel runs in production without writing a line of code. - What Flaky Tests Cost a 75-Engineer Team — https://mendral.com/blog/flaky-tests-cost A 75-engineer team loses about $375,000 a year to flaky tests in developer time alone. Here's the math, the CI compute bill, and the costs you can't put in a spreadsheet. - You should not update your dependencies in 2026 — https://mendral.com/blog/you-should-not-update A brief (irreverent) history of software supply chain security, and what to do about it in the age of AI. - Supply Chain Attacks Don't Wait for CVEs — https://mendral.com/blog/supply-chain-attacks-dont-wait-for-cves By the time a CVE exists, the bad version is already in thousands of CI pipelines. Here's the last year of npm and Actions attacks, why scanners are reactive by design, and the supply chain agent we shipped to catch dependency changes at the PR. - How We Know If Our Agent Is Right — https://mendral.com/blog/how-we-know-if-our-agent-is-right 36,564 investigations, no ground truth, and the eval signals we actually trust. - The Agent Harness Belongs Outside the Sandbox — https://mendral.com/blog/agent-harness-belongs-outside-sandbox Two architectures for running agent harnesses, the tradeoffs between them, and how we make skills and memories work when the harness isn't local. - Same LLM, Different Agent: What Changes When You Specialize for CI — https://mendral.com/blog/same-llm-different-agent Same models, different tokens. What changes when you build an agent harness specialized for CI instead of general-purpose coding. - We Upgraded to a Frontier Model and Our Costs Went Down — https://mendral.com/blog/frontier-model-lower-costs We switched to a frontier model and our costs went down. Here's the architecture that made it possible. - We use Claude Code daily. We still built our own CI agent. — https://mendral.com/blog/how-we-built-our-ai-agent How Mendral closes 16,000 CI investigations a month: three Anthropic tiers, Firecracker microVMs, durable execution on Inngest, and a custom Go agent loop. - LLMs Are Good at SQL. We Gave Ours Terabytes of CI Logs. — https://mendral.com/blog/llms-are-good-at-sql We gave our AI agent a SQL interface to billions of CI log lines in ClickHouse. How we ingest, store, and query 1.5 billion log lines a week. - What CI Actually Looks Like at a 100-Person Team — https://mendral.com/blog/ci-at-scale 575K CI jobs, 1.18 billion log lines, 33 million test executions in one week. What we learned building an AI agent for PostHog's CI at scale. ## Frequently asked questions ### What is Mendral? Mendral was the AI DevOps Engineer — three always-on agents for security, reliability, and performance, plus custom automations for any DevOps work specific to your stack. The team is joining Anthropic and the hosted product is winding down. ### Can I sign up for Mendral? No. Mendral is no longer onboarding new teams. See the announcement: https://mendral.com/blog/mendral-team-joins-anthropic ### Is Mendral the same as Mistral AI? No. Mendral (with an "e") is a Y Combinator-backed AI DevOps company building the AI DevOps Engineer. Mistral AI is a separate French foundation-model lab. The two companies are unrelated. ### What does Mendral actually do? Three always-on agents — Security, Reliability, and Performance — run from day one. The Security Agent reviews dependency PRs, pins safe versions, and surfaces only the CVEs actually exploitable in your code; it also catches compromised dependencies, malicious actions, and leaked secrets before they hit production. The Reliability Agent diagnoses CI failures and fixes flaky tests. The Performance Agent cuts build time via caching, parallelism, and slow-test pruning. On top of that, you can add custom automations triggered from CI/CD events, infra alerts (Datadog, Sentry, cloud deploys), schedules, Slack, Linear, or webhooks. ### How does Mendral install? Install the Mendral GitHub App, optionally connect Slack, and Mendral starts reading your CI logs immediately. First insights within minutes, first auto-fix within hours. Total setup time is under five minutes. ### Which CI providers does Mendral support? GitHub Actions today. Buildkite, CircleCI, and GitLab CI support is coming next. ### Who built Mendral? Sam Alba (Co-founder & CEO; ex-VP Eng at Docker, Dagger co-founder) and Andrea Luzzardi (Co-founder & CTO; ex-Docker, Dagger co-founder). Founded 2025. Y Combinator W26. ### Is Mendral SOC 2 compliant? Yes — SOC 2 Type II. See https://mendral.com/security for details. ### Where is Mendral based? San Francisco, California, USA. ### How can I contact Mendral? Email hello@mendral.com, or reach out on LinkedIn or X (@MendralHQ). ## Canonical links - Announcement: https://mendral.com/blog/mendral-team-joins-anthropic - Home: https://mendral.com - About: https://mendral.com/about - Security: https://mendral.com/security - Blog: https://mendral.com/blog - RSS: https://mendral.com/api/rss.xml - Sitemap: https://mendral.com/sitemap.xml - Y Combinator: https://www.ycombinator.com/companies/mendral - LinkedIn: https://www.linkedin.com/company/mendral/ - GitHub org: https://github.com/mendral-ai - GitHub App: https://github.com/apps/mendral-app - X / Twitter: https://x.com/MendralHQ - Status page: https://status.mendral.com