Background

Startups need to move fast, but DevOps and infrastructure management can slow them down. Gradient Labs, a startup building advanced AI agents, was founded by three ex-Monzo Bank engineers who wanted to avoid these bottlenecks. Neal Lathia, CTO, had extensive experience with Go at Monzo and was looking for a way to simplify infrastructure management so the team could focus on building their product.

The Challenge

Gradient Labs needed to move quickly with a small team and limited resources. They initially built prototypes in Python but preferred Go for its performance and reliability. The key challenges were:

• Automating infrastructure management
• Reducing DevOps overhead
• Enabling fast iteration and collaboration
• Ensuring scalability from day one

The solution

To solve for their challenges, they built a stack that uses:

• Backend framework: Encore.go
• Some AI-specific services written in Python
• Databases: Postgres with PGVector for AI
• Cloud hosting: Google Cloud Platform
• Infrastructure and DevOps automation: Encore Cloud
• Durable execution engine: Temporal

Adopting Encore Cloud for infrastructure and DevOps automation

Neal had experimented with Encore before and introduced it to the team. Encore provided an integrated backend framework that handled infrastructure, coupled with complete DevOps automation on GCP using Encor Cloud.

"DevOps automation was the biggest pull when selecting Encore, we didn’t want to waste time setting up deployments or managing infrastructure, and Encore solved that for us.

We also appreciated that it wasn't a separate DevOps tool and integrated with the code that we were writing.

We've been able to quickly build tooling that's made us even faster; right now at Gradient Labs when you open a PR, we consume an Encore Cloud webhook event when the preview environment is deployed and send a DM to the developer who opened the PR to say "your preview environment is ready for you"

We even do our deployments via Slack right now, we just click the button in Slack. That's just basically using Encore Cloud webhooks. It's just these like little things that helped build a lot of momentum for us."

Neal Lathia — CTO at Gradient Labs

Results and Impact

1. No Need for a DevOps Team

Encore automated infrastructure management, eliminating the need for dedicated DevOps engineers and allowing the team to focus on product development.

"We’re a team of 11, and none of us are DevOps engineers. That’s because Encore automates what we would have otherwise had to build ourselves."

Gradient Labs leveraged Encore’s built-in static analysis, webhooks, and preview environments to streamline their workflow. Deployments are now triggered via Slack, using Encore webhooks for notifications and automations.

2. Fast Developer Onboarding

Encore’s opinionated framework made it easy for new engineers to start contributing immediately. Every new hire at Gradient Labs commits and deploys their first PR on day one.

"Onboarding time is focused on learning our product, not figuring out how to deploy code. That’s a huge efficiency boost."

3. Scalability and AI-Specific Capabilities

Gradient Labs needed strong AI infrastructure support, including embedding and similarity search with PGVector in Postgres. Encore’s built-in Postgres support enabled seamless implementation.

"We use PGVector for AI similarity search, and Encore supports it natively. That made it an easy choice."

4. Seamless Integration with Other Tools

While Encore handles most of the backend, some AI-specific services are built in Python. Integration has been smooth, with webhooks triggering deployments across different environments.

"We open a PR in the agent repo, consume webhooks in Encore, and trigger deploys. It’s a well-established pattern, and Encore makes it easy."

Looking Ahead

Gradient Labs is exploring bringing on further languages as they grow their engineering team.

"Encore has been a huge help. The only real challenge we face is managing multiple languages, which isn’t specific to Encore, but something we need to figure out as we scale."

Key Takeaways

• Faster Time to Market – Avoided a multi-year DevOps buildout by using Encore from the start.
• Reduced DevOps Overhead – No need for dedicated DevOps engineers, freeing resources for product development.
• Rapid Developer Onboarding – New engineers push production code on day one.
• Scalability & AI Readiness – Built-in support for Postgres, PGVector, and AI workloads.
• Security Built-in – Simplified security with structured infrastructure management.

Conclusion

Encore enabled Gradient Labs to scale quickly without the complexity of DevOps. The combination of an opinionated framework, automated infrastructure, and AI-friendly capabilities positioned them for long-term success.

"If you're a startup, you don’t want to waste years struggling with DevOps before you can afford a platform team. Use a tool that handles it for you. Encore is a great choice."

Interested in seeing how Encore can work for your startup? Book a 1:1 demo to learn more.