Mendral

Mendral is The AI DevOps Engineer. Three always-on agents handle security, reliability, and performance, plus custom automations for anything else specific to your stack. From the team behind Docker and Dagger (YC W19). We run in production at PostHog, Metabase, Luminai, Inngest, Clipboard Health, Blaxel, among others. Across millions of CI jobs, 73% of failures auto-match to known issues. PostHog accepted 104 fixes from Mendral in a single month. The Role We're looking for an engineer who writes. Someone who can sit in our codebase, understand how the agents work, and publish the kind of technical posts that other engineers actually read and share. We're not hiring a marketer. We're hiring an engineer whose job is to make other engineers smarter about CI, agents, and DevOps. Pipeline follows from great content. We're not asking you to measure it, build funnels around it, or chase attribution. What you'll work on Technical blog posts that earn the front page of HN. Examples we've shipped: "We gave terabytes of CI logs to an LLM" (226 points), "What CI Actually Looks Like at a 100-Person Team," "Anatomy of a Production AI Agent." Conference talks. DevOpsDays, KubeCon, AI engineering events, anywhere senior engineers go to hear from peers. Meetups. Attending, speaking, hanging out with practitioners and bringing back what they're actually struggling with. Open-source contributions and code samples. Small, sharp repos that demonstrate ideas from our posts. Product demos. Recorded for the site, live for engineering teams who want a deeper look at the system. Customer engineering deep-dives. Working with teams at PostHog, Metabase, and others to write about real implementations at scale. Community presence on HN, Reddit, and X. Engaging engineer-to-engineer, not broadcasting. What we're looking for Engineering background. You've written code in production. You can read a CI log, reason about a flaky test, and write about it without translation help from us. Strong technical writer. You have public examples we can read. Blog posts, RFCs, postmortems, OSS docs. Speaker or aspiring speaker. You've given talks before, or you want to. Comfortable on camera and on stage without three weeks of prep. Curious about the stack. You find agent architecture, CI internals, and infrastructure interesting on their own terms, not as marketing surface. High agency. You're the first hire in this seat. Nobody is briefing you. You're the brief. Why now Seed-funded with named customers shipping fixes from Mendral every day. Founders who built Docker and Dagger and write engineer-first content themselves. We've already had multiple HN front-page posts. Your job is to make that the norm, not the exception. You'll be one of the first full-time hires, and the equity reflects that. Based in the SF Bay Area is preferred, or remote in a compatible timezone.

Mendral is The AI DevOps Engineer. Three always-on agents handle security, reliability, and performance, plus custom automations for anything else specific to your stack. From the team behind Docker and Dagger (YC W19). We run in production at PostHog, Metabase, Luminai, Inngest, Clipboard Health, Blaxel, among others. Across millions of CI jobs, 73% of failures auto-match to known issues. PostHog accepted 104 fixes from Mendral in a single month. The role We're looking for our a full-time engineer who thinks and operates like a co-founder. Someone who understands what we're building, why it matters, and can make high-quality technical decisions without waiting for direction. You'll own large surface areas of the product and infrastructure. You'll ship fast. You'll make architectural choices that pay off for years. The expectation isn't to write every line of code yourself. It's to make the right calls on what gets built, how, and why. What you'll work on Agent architecture and orchestration across hundreds of concurrent CI runs Multi-model routing across Anthropic and OpenAI, picking the right model per task The log ingestion pipeline (ClickHouse, billions lines of log ingested per week) that powers agent investigations Sandboxed execution, where each agent gets a full dev environment New agents (custom automations, security workflows, performance analyses) and the framework they share The GitHub App, Slack, and the integrations layer pulling signal from infrastructure (Datadog, Sentry, GCP, AWS, etc…) What we're looking for Builder mindset. You've shipped real products, not just features. You've made hard tradeoffs with incomplete information and lived with the consequences. AI-native workflow. You actively use AI coding agents every day. You understand prompt engineering, context management, and running multiple agent instances in parallel. Taste over volume. We care more about architecture decisions, system design, and technical judgment than lines of code. CI/CD and infrastructure fluency. Real experience with GitHub Actions, CI pipelines, containers, and cloud infrastructure. High agency. You enjoy a high level of autonomy and know how to use it. Why now Seed-funded with named customers in production. Founders with a decade of DevOps experience (Docker, Dagger). You'll be one of the first full-time hire, and the equity reflects that. Based in the SF Bay Area is preferred, or remote in a compatible timezone.