Executive summary
Dexlyn wanted business users to understand what drove revenue and other key metrics without waiting on technical teams. Each answer meant manual SQL and combining results by hand. DevOps TechLab deployed QueryDash, its agentic investigation platform built on Amazon Bedrock, as an autonomous business intelligence agent: users ask the question in plain language, and the agent works through the data step by step and explains its findings with evidence.
- Client
- Dexlyn · Business intelligence
- Business problem
- Explaining what moved revenue and key metrics needed manual SQL and hand-combined results.
- What we delivered
- An autonomous BI agent (QueryDash) on Amazon Bedrock with Amazon Nova Pro, connected read-only to Dexlyn’s business data.
- Engagement
- Discovery, build and onboarding, then managed operations and continuous optimisation by DevOps TechLab.
- Outcome
- Business users get multi-step, evidence-based answers to questions about revenue and performance.
The business challenge
Understanding what drove revenue and other key metrics meant writing SQL manually and combining results by hand. Complex business questions took longer to answer than the business wanted, and answers depended on technical expertise.
Why it mattered
- Slow answers to strategic questions
Questions about what moved revenue waited on manual analysis.
- Dashboards stop at “what”
Fixed reports showed that a metric changed, not which factors caused it.
- Results combined by hand
Several queries had to be run and their results joined manually to see the full picture.
- Expertise as a bottleneck
Only people with SQL skills could investigate, so business users waited for them.
What Dexlyn set out to achieve
- Answer business questions such as revenue drivers in plain language.
- Rank the factors behind a change, backed by evidence.
- Let users ask follow-up questions in the same investigation.
- Keep production data safe from AI-generated queries.
- Extend to new business questions without new analytics projects.
The solution: an autonomous business intelligence agent
Users ask a business question, for example “Determine the main factors affecting quarterly revenue.” The agent plans a multi-step investigation, discovers the relevant data, generates and validates each query, analyses the results and explains which factors matter most, recovering on its own when a query fails.
Understand the goal. The agent restates what you are trying to find out.
Plan. It breaks the question into steps.
Discover the data. It finds the tables and columns that matter, and how they relate.
Generate SQL. It writes the SQL for the next step.
Validate: read-only. Every statement is checked before it runs. Writes and schema changes are rejected.
Execute. The query runs against production data, read-only.
Analyse. It reads the results and decides what they mean.
Replan or retry. If a query fails or the answer is incomplete, it adjusts the plan and tries again, until it can explain the answer.
Key design choices
- Managed generative AI
Amazon Bedrock with Amazon Nova Pro handles reasoning and tool use. There are no models to host and no GPU infrastructure to run.
- No servers to manage
The application runs on Amazon ECS on AWS Fargate behind an Application Load Balancer, so it scales with demand.
- Works on the data where it lives
The agent queries the existing operational databases directly, read-only, instead of copying data into a new warehouse.
- Investigations that remember
Amazon DynamoDB keeps the conversation and investigation history, so follow-up questions build on what the agent already found.
Security and governance
For leadership, the first question about AI on production data is risk. The platform is built so that AI-generated queries never touch the database directly.
- Read-only by design
Only read statements (SELECT, WITH, SHOW, DESCRIBE, EXPLAIN) are allowed. Any write or schema change is rejected before it runs.
- Validated before execution
Every query the AI writes passes through a server-side validation layer. The model never has a direct line to production data.
- Protected performance
Results are capped, so a broad question cannot turn into a runaway query on a production database.
- Managed credentials
Database credentials sit in AWS Secrets Manager, and access is governed by AWS IAM roles, not stored in code.
Architecture on AWS
| AWS service | Role in the solution |
|---|---|
| Amazon Bedrock · Amazon Nova Pro | Agent reasoning, planning and tool use |
| Amazon ECS on AWS Fargate · Application Load Balancer | Runs the investigation service without servers to manage |
| Amazon RDS (MySQL) · Amazon Aurora PostgreSQL | The business data the agent investigates, read-only |
| Amazon DynamoDB | Conversation and investigation memory |
| Amazon CloudFront · Amazon S3 | Delivers the web application securely |
| AWS Secrets Manager · AWS IAM | Credentials and access control |
| Amazon CloudWatch | Application, infrastructure and agent monitoring |
How we delivered
- 01Discover
Mapped the questions Dexlyn’s users need answered, the data behind them and how it relates, and defined secure, read-only access to production.
- 02Build & onboard
Connected Dexlyn’s data, configured Amazon Bedrock and Nova Pro, set up read-only validation, conversation memory and real-time streaming, and tested against representative investigation scenarios.
- 03Run
DevOps TechLab operates the service: monitoring the application, the AWS environment and the agent, supporting queries and resolving production issues.
- 04Improve
Ongoing tuning of agent planning, query generation and recovery, plus reviews of performance, cost and security as new questions are added.
Business outcomes
Dexlyn’s business users now get evidence-based answers about what drives revenue in one conversation, instead of waiting for manual analysis.
- Self-service business intelligence
Business users investigate revenue and key metrics in plain language, without SQL.
- Drivers, not just numbers
The agent works through the data to identify the factors behind a change.
- Evidence-based findings
Every answer comes with the analysis behind it, so leaders can act with confidence.
- Zero write access to production
AI-generated queries are validated and read-only.
- A platform, not a one-off
New business questions are added on the same platform, without a separate analytics project for each.
What’s next
Because the architecture is reusable, Dexlyn can extend the agent to new business questions without building a separate analytics workflow for each. DevOps TechLab continues to advise on new investigation scenarios and further opportunities for AI-driven automation.
The same platform, other teams
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 answers like these from your own data?
Share the questions your leadership team asks most often. We will show how an AI investigation agent on Amazon Bedrock could answer them, safely and read-only.
