How NaNLABS helped Delos stabilize their platform, automate key workflows and scale without hiring internally

How NaNLABS helped Delos stabilize their platform, automate high-impact workflows, and scale without growing internal headcount

Sharing is caring!

by Matias Emiliano Alvarez Duran

06/25/2025

Delos Insurance is redefining how wildfire coverage works in high-risk areas. But behind their groundbreaking data model was a platform held together by manual work, technical debt, and hiring bottlenecks.

Here’s how NaNLABS embedded into their team, replaced fragile processes with automation, and helped turn a reactive dev org into a scalable delivery machine.

Meet the client: Wildfire insurance, powered by data

Delos is a San Francisco-based Managing General Agent (MGA) that uses satellite imagery, climate analytics, and proprietary modeling to offer home insurance in wildfire-prone zones, areas most carriers won’t touch.

By building smarter risk models, Delos expands coverage to high-risk homeowners while controlling exposure for carriers. Their secret weapon? Tech.

But in 2023, the gap between business momentum and tech infrastructure began to show. This is a company where the technology followed the business. So there’s a lot of technical debt.”, said Amrit Williams, Delos CTO.

The brief: Stabilize, automate, and grow faster

When Delos reached out to NaNLABS, they weren’t just looking for devs. They needed:

  • Engineers who could embed in their core product team quickly

  • Automation to replace spreadsheet-heavy operations

  • QA and CI/CD support to reduce release anxiety

  • Flexibility to scale without hiring FTEs in the U.S.

“I tried hiring one person and it took weeks… I just didn’t have the time or capacity for that.”, confessed Williams. The relationship started with backend work, but soon expanded to QA automation, DevOps, cross-team support, and more.

The challenge: Fragile tech, no QA, manual ops

Delos had product-market fit and carrier demand, but internally, three core problems were holding them back.

1. Manual workflows that couldn’t scale: Core business functions like agent commission payouts and SL2 form submissions for regulators were being managed with spreadsheets and manual Stripe entries.

2. QA and release bottlenecks: With no QA team and no test automation in place, developers had to test their own code. There were no reliable regression suites or CI guardrails. Every deployment felt risky.

3. Technical debt and brittle integrations: Each new insurance program had its own rules and third-party integrations, some even excluding homes with solar panels. The growing complexity of logic and APIs strained their backend. Add to that infrastructure knowledge siloed in one contractor, and the team faced real risk of slowdowns, outages, or burnout.

The solution: Embedded engineers with a sidekick mindset

Rather than deliver code from the outside, NaNLABS embedded across Delos’ team, sharing their backlog, standups, and business context. We approached each challenge with long-term thinking, asking not just “what needs doing?” but “how can we make this easier next time?”

Here’s how we helped:

Backend and Full-Stack Engineering

Our team of backend and full-stack engineers focused on unblocking core business flows and stabilizing the platform:

  • Built reusable APIs for commission payouts and SL2 form generation

  • Refactored brittle logic to support new insurance program rules

  • Improved modularity and maintainability across backend services

QA Automation

We introduced the first formal QA process at Delos by:

  • Implementing Cypress for end-to-end test automation

  • Collaborating with product to define test ownership and coverage goals

  • Building a regression suite to reduce release fear and manual rework

DevOps & CI/CD

To reduce deploy friction and improve reliability, we:

  • Refactored GitHub Actions pipelines

  • Reduced reliance on external contractors for infrastructure knowledge

  • Scoped infrastructure observability improvements for the next phase

Cross-functional engineering support

Delos is a lean team, so when our engineers saw inefficiencies outside the product squad, like billing or operations logic, we stepped in.

  • Supported billing automation in collaboration with the Ops team

  • Suggested improvements to internal workflows not covered in the original brief

“Your folks aren’t just coders. They’re jumping into billing and ops when needed. That’s rare.” said Williams, praising our proactive sidekicks. 

Results: Automation, velocity, and long-term trust

The outcome of our partnership wasn’t just working features, it was a more resilient, agile, and confident product org.

Impact AreaBefore NaNLABSAfter NaNLABS
SL2 WorkflowsManual spreadsheets & Stripe exports Fully automated API-driven pipelines
QA CoverageManual testing, no regression suite Cypress automation + defined QA process
CI/CD & Deploys Fragile pipelines & single-point riskCleaned pipelines + shared infra context
Backend ScalabilityFragile logic, inconsistent patternsModularized APIs, carrier-ready backend
Cross-team CollaborationEngineering-only scopeEmbedded across Ops, Billing, QA

Delos didn’t just gain velocity. They gained a partner who helped stabilize their platform, own complex backend logic, and reduce the risk and cost of scaling, all without expanding internal headcount.

"
"NaNLABS is the best company I've ever worked with for South American tech talent. Their team is curious, they ask why, they dig deep, they deliver quality ... and they fit right in. I trust them and I'm already introducing NaNLABS to other companies" - Amrit Williams, CTO, Delos Insurance

Thinking of team augmentation? Start with trust

At NaNLABS, we don’t just fill gaps, we become your tech sidekick. We embed quickly, collaborate deeply, and help you move from fragile code and manual pain to scalable infrastructure and lasting impact.

If you're facing similar challenges, technical debt, release risk, hiring fatigue, let’s talk.


More articles to read

Previous blog post

Web Technologies

11/18/2013

Avoiding huge controllers in AngularJS

Read the complete article

Next blog post

Web Technologies

06/18/2025

Business Impact of EV Charging Stations

Read the complete article