Case Studies | SaaS B2B

Migrating a battery intelligence platform to AWS

AWS Migration

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge spanned four focus areas that the existing Vercel and Firebase hosting could not fully address.

Real-time telemetry and time-series at fleet scale

The platform needed to ingest Battery Management System telemetry from thousands of packs in real time and retain per-battery time-series metrics such as voltage, current, temperature, state of charge and state of health, well beyond what the incumbent hosting was built for.

Enterprise identity and access

Selling to enterprise battery manufacturers required enterprise SSO and SAML federation, MFA and a full role-based access model, replacing Firebase Auth with an identity layer that enterprise procurement would accept.

AI inference and GPU-based ML

The platform, the platform narrative generation and risk analytics needed a primary inference path, and state-of-health scoring and physics simulation needed GPU compute for batch prediction, model retraining and electrochemical modelling.

Regulatory chain of custody

The platform compliance under EU Battery Regulation 2023/1542 demanded encrypted, auditable chain-of-custody records, long-term data retention and EU data residency, none of which the original architecture was structured to guarantee.

Solution

Cloud Combinator delivered the migration in phases, cutting each application over with a blue/ green strategy at the environment-variable level so that active users saw no downtime, with rollback preserved throughout.

19

Next.js applications migrated to Amazon ECS Fargate

80+

Firestore collections migrated to Amazon DynamoDB

3,000

Battery packs on the telemetry pipeline at initial fleet scale

By the numbers:

  • 19 - Next.js applications migrated to Amazon ECS Fargate
  • 80+ - Firestore collections migrated to Amazon DynamoDB
  • 3,000 - Battery packs on the telemetry pipeline at initial fleet scale
Changes

Acceptance is the joint confirmation that all success criteria are met, that Vercel and Firebase are fully decommissioned, and that the platform operates end to end on AWS in eu-west-1 with no dependency on the former hosting. The figures below describe the scope of the migration delivered.

  • Fully re-platformed onto AWS — All 19 applications run on ECS Fargate behind an ALB in eu-west-1, with Vercel deployments and Firebase services decommissioned once production traffic is validated on AWS.
  • A regulator-ready data foundation — IoT Core, Timestream and a S3 Parquet data lake deliver real-time telemetry with a seven-year cold retention path, and CloudTrail plus KMS provide the encrypted chain of custody the platform requires.
  • AI and ML on AWS-native services — Amazon Bedrock becomes the primary inference path while SageMaker and AWS Batch cover fleet-wide state-of-health scoring, model retraining and physics simulation.
  • Enterprise identity in place — Amazon Cognito provides enterprise SSO and SAML federation, MFA and the full role-based access model expected by enterprise buyers.
  • Handover — An architecture handover pack with IaC repositories, operational runbooks and FTR readiness documentation is delivered, alongside AWS Marketplace listing configuration and ACE pipeline registration.

With the platform running end to end on AWS, the client is positioned to scale from its initial fleet toward the enterprise customer base and battery volumes it is targeting, with the regulatory foundations for the platform compliance already in place.

AWS Stack

Amazon ECS Fargate

For running all 19 containerised Next.js applications behind an Application Load Balancer.

Amazon DynamoDB

For the multi-tenant data store, migrated from Firebase Firestore with GSIs and point-in-time recovery.

Amazon Cognito

For enterprise SSO, SAML federation, MFA and role-based access control.

AWS IoT Core

And Amazon Timestream for real-time battery telemetry ingestion and time-series storage.

Amazon Bedrock

(Claude 3.5 Sonnet) for primary AI inference across the platform, the platform and risk analytics.

Amazon SageMaker

And AWS Batch for state-of-health scoring, model retraining and BattMo physics simulations.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.