Caddi chooses Gruntwork Pipelines over Atlantis

Key results:

• Setup in under an hour: Pipelines went from zero to deployed infrastructure on day one.
• No more infra babysitting: once live, configs stayed untouched and stable.
• Logs where engineers work: plans and applies appeared directly in GitHub pull requests.
• Consistent deployments: every merge deployed automatically, eliminating drift.
• More focus on product: time once spent patching pipelines shifted to building features.

‍

Caddi Technologies is a fast-growing automation platform that turns screen-recorded workflows into push-button automations. Securing seed funding in February 2025, the company is growing fast and building out its engineering team and infrastructure.

As a founding engineer, Dallas Slaughter wears many hats but comes from a DevOps background. He quickly saw that the way Caddi deployed infrastructure (e.g. running terraform apply locally) wouldn’t scale. Debugging was painful, visibility was low, and engineers often needed him to step in to help troubleshoot.

Slaughter knew the tooling landscape well. In prior roles he had evaluated or ran Atlantis, Spacelift, Terrateam, and GitHub Actions setups. He’d spent years administering Atlantis, only to find it heavy, slow, and fragile, especially once Terragrunt was involved. For him, managing Atlantis “was almost a part-time job.”

This time, he wanted a clean start.

“Atlantis plus Terragrunt adds a big chunk of work. Once you add Terragrunt in the mix (and I can’t imagine doing IaC without Terragrunt at this point) it’s a no go.” – Dallas Slaughter, Founding Engineer at Caddi

Starting on the right foot with Gruntwork Pipelines

Slaughter had no interest in repeating past frustrations. He wanted a new system that wouldn’t buckle under the weight of large Terragrunt repos, stall for nearly an hour, or erase logs the moment an ECS task cycled. Instead of constant patching, security updates, and configuration tweaks — work that offered no real business value — he turned to Gruntwork.

Gruntwork offered exactly what Caddi needed: a simple setup, seamless support for Terragrunt, and a deployment flow that matched how the team already worked in GitHub. He decided to run all the infrastructure management through Pipelines, giving the entire company visibility into changes and the confidence that what lived in the repo was what ran in production. 

It was the clean foundation Slaughter had been looking for:

• Faster onboarding: Slaughter didn’t burn any cycles on setup. He had Pipelines live in under an hour, meaning his team could ship infrastructure right away instead of waiting on a drawn-out tooling rollout.
• Less overhead with Terragrunt: Slaughter no longer had to wire up custom logic or watch slow, full-repo plans crawl through. Pipelines handled Terragrunt repos natively, so the team spent time building features instead of maintaining workarounds.
• Shared visibility in GitHub: engineers could see logs and apply history alongside their pull requests. When something broke, Slaughter didn’t have to play translator or troubleshooter since anyone on the team could trace the issue back to the exact commit.
• Confidence in deployments: the team no longer had to second-guess whether an environment matched what was in code. Merging a PR meant it was deployed, which eliminated “do not touch” modules and stale infra.
• Time back for engineering: with no config to babysit after setup, Slaughter focused his energy on Caddi’s product instead of patching tooling. The team moved faster without pulling him into every troubleshooting session.

“Leaning into Gruntwork products has always been good for me. They’ve always done me really well and made my life easier in the end too. So it was very natural to reach for it.” – Dallas Slaughter, Founding Engineer at Caddi

Implementation and outcomes: from setup to impact

Working directly with Gruntwork, Slaughter had Pipelines fully operational in under an hour. Slaughter jokingly recalls, “We were on a call for 30 or 45 minutes, and we were up and running.” 

The initial onboarding session went from zero to deployed infrastructure. Slaughter adds, “I haven’t touched the config once since we set it up. It’s just there. It just works.” GitHub Actions handled every apply with centralized logs, merges translated directly into deployed state, and Pipelines intelligently applied only what changed along with the right dependencies. (Even better, Pipelines runs entirely in your GitHub Actions environment, meaning Gruntwork can’t access your state files or secrets.)

Pipelines also introduced a discipline that shaped how the team worked. Every merge resulted in a deployment, full stop. While that left less room for shortcuts, it gave Caddi the assurance that what lived in code was exactly what ran in production. Over time, that consistency became part of the team’s rhythm and removed the uncertainty that often creeps in when environments drift apart.

Even when something went wrong, the difference was clear. The only issue (a lock collision when env-common and a module applied at the same time, scheduled to be addressed in an upcoming release) was obvious to diagnose and fixed by rerunning the job. By contrast, Slaughter recalled Atlantis turning similar hiccups into multi-hour hunts across web UIs and missing logs. With Pipelines, troubleshooting was quick and transparent, keeping engineers focused on building instead of firefighting.

The day-to-day effect was simplicity: after setup, Pipelines just worked. Aside from the occasional minor lock collision—resolved by rerunning the job and already on Gruntwork’s roadmap—the system required no maintenance. That freed Slaughter to focus his time on building Caddi’s product, while the rest of the team gained confidence and velocity without dragging him into every troubleshooting session.

“We went from local, opaque applies to a shared, auditable pipeline in under an hour.” – Dallas Slaughter, Founding Engineer at Caddi

Caddi can build on a stronger foundation

With Pipelines in place, Caddi has a foundation it can rely on as the company scales. The team no longer worries about drift, fragile tooling, or maintenance chores that distract from product work. As infrastructure needs grow more complex, Pipelines provides the consistency and visibility to keep delivery on track.

The value goes beyond this rollout. Slaughter has turned to Gruntwork at three or four companies already and says he would do the same again. That track record underscores a level of trust that comes only from seeing consistent results over time.

For Caddi, the payoff is clear: every engineer’s time counts, and Pipelines gives them more of it. By eliminating babysitting and patching, Slaughter spends more of his energy coding features, while the rest of the team can build and deploy infrastructure without slowing product delivery. Gruntwork has given Caddi both the technical footing and the business efficiency to keep pace with its growth.

Learn how Gruntwork Pipelines can replace Atlantis and give your Terragrunt workflows the CI/CD layer they deserve.

‍