TCS built a BFSI agentic test orchestrator on Bedrock AgentCore with per-tenant isolation
AWS ML Blog · rss · 2026-11-11
An AWS ML Blog engineering writeup details how TCS built an Agentic Test Orchestrator on Amazon Bedrock AgentCore for BFSI clients, targeting the slow parts of releases: requirements interpretation, scenario design, script maintenance, and failure triage — not replacing QA engineers. Key design points:
- Specialized agents (scenario generation, Gherkin conversion, Java/Python codegen, validation, chat refinement, BRD-to-story, execution) orchestrated by a deterministic supervisor, with retries, error classification, and reporting in a dedicated workflow
- Context grounding via Bedrock Knowledge Bases: Jira stories, BRDs, Swagger specs, GitLab source, and Puppeteer-captured UI flows, retrieval filtered per tenant/project/session
- Multi-tenant isolation: shared control plane plus per-tenant Cognito pools, KMS keys, S3 buckets, DynamoDB tables; event-driven CDK onboarding via EventBridge
- Agents built with Strands Agents SDK, each session in an isolated microVM; tests run on Amazon ECS
More from coding & agent
- Mitra adds MCP support so ChatGPT and Claude can make calls from your phone number — testingcatalog · 2026-11-11
- LangChain: evidence-graph GraphRAG lifts multi-source agent accuracy from 10% to 52% — JeremyCMorgan · 2026-11-11
- AI agents are finding critical CVEs — design your SaaS assuming the server is lost — haltakov · 2026-11-11
- Workday agent prototyping cut from 4-5 months to under a day with LangGraph — LangChain · 2026-11-11
- Unreal Engine 5.9 Preview is the first engine release built mostly by AI agents — nicolascraske · 2026-11-11
- Figma MCP users beg for read:comments — fetching comments still needs a PAT — voidwalker_com · 2026-11-11