Awaiting customer approval. This page isn't marketed anywhere.
AheadComputing
Location
Portland, Oregon, USA
Customer Since
Apr 11, 2025
Team Size
100 engineers
Ben Holtzman
Senior Design Enablement Engineer
AheadComputing designs breakthrough 64-bit RISC-V application processors, targeting next-generation workloads across AI, cloud, and edge devices.
Complex hardware CI pipelines with long-running tests (up to 90 minutes) slowed down development feedback loops.
GitHub’s native merge queue couldn’t handle Ahead’s advanced requirements — freezes, prioritization, and custom rules.
Lack of CI observability made it hard to identify wasted compute and flaky test behavior.
AheadComputing is building a next-generation RISC-V CPU core, created by a team of former Intel engineers with deep experience in high-performance compute. As the company grew, its CI system became a bottleneck.
"Hardware CI is a different world," explains Ben Holtzman, Infrastructure Engineer. "Tests can run for 90 minutes, and you can't afford false failures or mismanaged queues.”
When Ahead first adopted GitHub's native merge queue, they quickly encountered scaling limits: a lack of control over scheduling, prioritization, and fine-grained role conditions. "We needed something programmable that could adapt to how we actually work," says Holtzman.
With Mergify, we can finally run our merge queues the way we want: custom rules, multiple queues, and fine-grained control over every condition.
Ben Holtzman
Senior Design Enablement Engineer
Ahead Computing fully migrated to Mergify's Merge Queue, using advanced scheduling and conditions to ensure reliable merges across multiple hardware and software workflows.
By combining scheduled freezes, multi-queue setups, and dynamic labeling, the team can manage merges for long-running tests without risking broken main branches or wasted compute.
They also implemented Mergify's CI Insights, gaining full visibility into runtime performance, skipped jobs, and resource usage across their monorepo.
Mergify gives us the control and visibility we never had with GitHub’s queue. It's become our CI command center.
Ben Holtzman
Senior Design Enablement Engineer
AheadComputing’s next step is to integrate flaky-test detection and automatic retries for self-hosted runners. “Our hardware tests are unique,” says Holtzman. “If we can automatically detect and retry flaky runs, that’s a massive time saver.”
By pairing Mergify’s Merge Queue with CI Insights, AheadComputing is creating a continuous integration system that scales with the complexity of hardware — without slowing engineers down.
