Case Studies

A colorful selection of our recent work and outcomes across data, APIs, DevOps, and agentic systems.

Multi agentic systems in AWS with Agentic Core

Design and operate multi-agent systems on AWS using Bedrock AgentCore Runtime: internal A2A, Strands-powered human approvals, Cognito/IAM identity, MCP tools via gateway (FastMCP), and observability with CloudWatch, OpenTelemetry, and Langfuse; deployed via GitHub Actions.

  • Short-lived, scoped tokens per agent
  • Human approvals via Strands SDK/Runtime
multi-agent systems, identity, cloudAWS Bedrock AgentCore Runtime, AWS SQS, AWS DynamoDB, AWS Cognito, IAM Identity Center, MCP Gateway, FastMCP, Strands SDK, Strands Runtime, GitHub Actions, OpenTelemetry, Amazon CloudWatch, Langfuse, JSON Web Tokens, OpenID Connect, IaC

AWS Datalake with Apache Iceberg

Incremental cross‑cloud ingestion from GCP BigQuery into an AWS data lake backed by Apache Iceberg.

  • Incremental ingestion via Iceberg snapshots
  • Schema evolution & partition management
data engineering, cloud, datalakeAWS S3, AWS Glue, Apache Iceberg, GCP BigQuery, BigQuery Spark Connector, Apache Spark

Azure API Management at Scale

Enterprise API platform on Azure APIM with global security, reusable policies, multi‑API support, and self‑hosted gateways for low latency, delivered via GitHub CI/CD.

  • Global security policy (authN/Z, rate limit)
  • Reusable policy fragments (transform, headers, caching)
api management, integration, cloudAzure API Management, APIM Policies, Self‑hosted Gateway, Azure Key Vault, Azure Monitor, GitHub Actions, OpenAPI, ARM/Bicep, APIOps (GitOps)

MCP Server Productionizing on AWS (ECS + FastMCP)

Harden and scale MCP servers on AWS ECS with FastMCP, ALB, and observability.

  • Zero-downtime deployments
  • Autoscaling policies & burst handling
agent ops, cloud, mcpAWS ECS Fargate, AWS ALB, AWS ECR, AWS CloudWatch, AWS IAM, AWS VPC, AWS Secrets Manager, FastMCP, MCP