Executive summary
The One Technologies handles a growing volume of corporate gifting enquiries. Every enquiry meant finding products, building hampers, checking prices and formatting a branded quotation by hand, which slowed responses and made proposals inconsistent. DevOps TechLab built a fully serverless AI platform on Amazon Bedrock that produces budget-compliant recommendations and branded PDF quotations within minutes, with no AI infrastructure for the company to run.
- Client
- The One Technologies · Corporate gifting
- Business problem
- Manual quotation work slowed responses and made proposals inconsistent as enquiry volumes grew.
- What we delivered
- AI recommendations, automated hamper generation with business-rule checks, and branded PDF quotations, on Amazon Bedrock.
- Engagement
- Discovery, design, build and security, then ongoing optimisation by DevOps TechLab.
- Outcome
- Branded, budget-compliant quotations within minutes and faster responses to customers.
The business challenge
Every customer request required employees to identify products, assemble suitable hampers, validate pricing and prepare a professionally branded quotation. As volumes increased, this became a bottleneck that slowed customer response times and introduced inconsistencies across proposals.
Why it mattered
- Slower responses to customers
Quotation work queued up as enquiry volumes grew.
- Inconsistent proposals
Quotes depended on who prepared them, so quality and format varied.
- Sales time spent on formatting
Valuable sales time went into building hampers and formatting documents.
- AI without the overhead
The company wanted AI, but not dedicated AI infrastructure to run.
What The One Technologies set out to achieve
- Generate product recommendations automatically from customer requirements.
- Build hampers that respect business rules and budgets.
- Produce consistent, branded quotations without manual formatting.
- Respond to customers faster as volumes grow.
- Run with no dedicated AI infrastructure.
The solution: AI quote and proposal automation
Sales staff enter the customer’s requirements. The platform recommends suitable products, generates a hamper that passes business-rule and budget checks, and produces a branded PDF quotation stored securely in Amazon S3.
Find products
→AI recommendations
Build hampers
→Hamper generated with business-rule checks
Check prices
→Budget checked automatically
Format a branded quotation
→Branded PDF stored in S3
Key design choices
- Managed AI on Amazon Bedrock
Chosen to keep operations simple: no models to host and nothing for the company to maintain.
- Fully serverless
Amazon API Gateway, AWS Lambda and Aurora PostgreSQL Serverless v2 scale with demand.
- Costed before the build
The architecture was reviewed with cost estimates before any build started.
- Documents stored securely
Branded PDF quotations are generated and stored in Amazon S3.
Security and governance
Customer data and pricing stay inside a private AWS environment, defined entirely as code.
- Least-privilege access
AWS IAM roles give each component only the access it needs.
- Secrets managed
Credentials are kept in AWS Secrets Manager, not in code.
- Private networking
The database sits on private networking behind Amazon RDS Proxy, with VPC endpoints for AWS services.
- Infrastructure as code
Everything is provisioned with AWS CloudFormation, so the environment is repeatable and reviewable.
Architecture on AWS
| AWS service | Role in the solution |
|---|---|
| Amazon Bedrock | AI product recommendations |
| Amazon API Gateway · AWS Lambda | Serverless application, hamper and quotation logic |
| Amazon Aurora PostgreSQL Serverless v2 · Amazon RDS Proxy | Product and quotation data, private access |
| Amazon S3 | Generated PDF quotations |
| AWS IAM · AWS Secrets Manager · VPC endpoints | Access control, secrets and private networking |
| AWS CloudFormation | Infrastructure as code |
How we delivered
- 01Discover
Mapped the quotation workflow from enquiry to final quote, and chose Amazon Bedrock to keep operations simple.
- 02Design
A fully serverless architecture on API Gateway, Lambda, Aurora Serverless v2 and Bedrock, reviewed with cost estimates before the build.
- 03Build & secure
AI recommendations, automated hamper generation with business-rule checks and branded PDF quotations, on private, least-privilege infrastructure.
- 04Optimise
Serverless cost tuning, Aurora tuned to the workload and faster document generation.
Business outcomes
Quotation work that used to be done by hand for every enquiry is now automated from request to branded PDF.
- Quotations within minutes
Branded, budget-compliant quotations are produced within minutes.
- Consistent proposals
Every quote follows the same rules and branding, whoever prepares it.
- Less manual effort
Sales staff spend less time finding products, building hampers and formatting documents.
- Faster responses to customers
Enquiries get answered sooner as volumes grow.
- No AI infrastructure to run
The platform is serverless and uses managed AI on Amazon Bedrock.
What’s next
The roadmap includes Amazon Cognito for user management and further automation of the quotation process. DevOps TechLab continues to optimise the platform as volumes grow.
DevOps TechLab is an AWS Advanced Tier Services Partner with the AWS Generative AI Competency and AWS SMB Competency. We design, build and run cloud and generative AI solutions on AWS for growing businesses, with 265+ cloud environments under management.
Want AI like this working for your sales team?
Tell us where quotes, searches or proposals slow your team down. We will suggest a small first build on Amazon Bedrock, with the running cost estimated up front.
