Senior Software Engineer - Release Engineering

Everlaw
Oakland, California, United StatesPosted 5 May 2022

Job Description

<p>Everlaw is evolving its release infrastructure to support a future where AI agents actively participate in deployments. We are looking for a hands-on Senior Software Engineer to build, optimize, and harden the pipelines, data models, and APIs that make this possible.</p> <p>This is an execution-heavy, software engineering role, not a “keep the lights on” position maintaining legacy tooling. You will be writing code, tuning Gradle/Bazel builds, debugging CI flakes, and implementing the specific API endpoints that allow both humans and agents to interact with our infrastructure safely. You will build the high-performance plumbing that powers our transition to an agent-ready platform.</p> <p>At Everlaw, engineering is central to our mission of promoting justice by illuminating truth. Our customers rely on us to handle mission-critical, highly confidential data in commercial and FedRAMP-regulated environments, and our release infrastructure is a key part of that trust.</p> <p>This is a full-time, on-site position located in Oakland, California with an expected hybrid schedule of 3 days/week in the office and 2 days/week at home.</p> <h3>Getting started</h3> <ul> <li><strong>We want you to feel like part of the team early on!</strong> Our onboarding process will integrate you into the company with informative sessions on our product, policies, processes, and team structure and goals. </li> <li><strong>We’re excited for you to learn, grow, and contribute right away!</strong> We trust that you’ll bring experience and knowledge that will uplift and uplevel the team, but we don’t expect you to know everything on Day 1.</li> </ul> <h3>About the Release Team</h3> <p>The Release team in Platform Engineering owns the “path to production” for the entire Everlaw engineering organization. We are responsible for:</p> <ul> <li>The Build: Managing compilation and packaging for our large JVM monorepo and related services.</li> <li>The Pipeline: Operating the CI/CD systems that validate and promote changes across FedRAMP and commercial environments.</li> <li>The Future: Making the release platform “agent ready.” That means refactoring our infrastructure to be deterministic, API-driven, and structurally transparent so AI agents can reason about changes without hallucinating, and can act within well-defined safety guardrails.</li> </ul> <p>You’ll work closely with Product Engineering, Security Engineering, and our emerging AI/Agentic DevEx efforts to make shipping changes to Everlaw faster, safer, and more automatable.</p> <h3>In your role, you'll...</h3> <h3>Modernize and Scale Core CI/CD and Build Performance</h3> <p>Your first mandate is to make our existing release platform fast, reliable, and observable.</p> <ul> <li>Reduce build and test times: Dive deep into Gradle/Bazel configurations and our Java build graph to improve cache hit rates, parallelism, and resource utilization for our monorepo.</li> <li>Increase signal quality: Systematically identify and eliminate flaky tests and unstable infrastructure in our CI pipelines so that a “green build” is a highly trustworthy signal.</li> <li>Automate compliance: Build automations that generate FedRAMP and other audit artifacts from our pipelines (e.g., change logs, approvals, evidence of controls), reducing manual toil while improving our compliance posture.</li> <li>Measure what matters: Instrument CI/CD and release workflows with metrics and dashboards for latency, reliability, cost, and failure modes, and drive concrete improvements from those insights.</li> </ul> <h3>Evolve Toward Automated Agentic Workflows</h3> <p>Full AI agent integration is a longer-term goal. As we mature our release platform, you’ll help lay the groundwork for “agent-ready” infrastructure by:</p> <ul> <li>Improving data hygiene: Gradually shifting logs and build outputs from unstructured text toward structured, traceable formats, making debugging easier for humans now and enabling future automation.</li> <li>Encouragin ... (truncated, view full listing at source)