whchoi98

Guides migration of code between Arm SoCs with architecture-aware analysis and safe migration practices. Use when mentioning Arm SoC, Cortex migration, embedded migration, or Arm architecture porting.

Build, test, and deploy AI agents using AWS Bedrock AgentCore with local development workflow. Use when mentioning AgentCore, Bedrock agents, or AI agent deployment on AWS.

Build full-stack apps with AWS Amplify Gen 2 using TypeScript, guided workflows, and best practices. Use when mentioning Amplify, Amplify Gen 2, fullstack AWS app, Cognito auth, or GraphQL backend.

Monitor AWS with CloudWatch and CloudTrail without MCP. Use when the user asks about logs, metrics, alarms, or API audit trail.

Analyze AWS costs, billing, and pricing without MCP. Use when the user asks about cost analysis, cost forecast, pricing lookup, or billing reports.

Query and manage AWS data services (DynamoDB, Aurora, Redshift, ElastiCache, Neptune, S3 Tables) without MCP. Use when the user asks about databases, caching, data queries, or data management.

Analyze source code for Graviton (Arm64) compatibility, identify issues, and suggest fixes for language runtimes and dependencies. Use when mentioning Graviton, Arm64 migration, x86 to Arm, or aarch64 porting.

Create, migrate, run, debug and optimize genomics workflows in AWS HealthOmics. Use when mentioning HealthOmics, WDL, CWL, Nextflow, genomics, or bioinformatics pipelines.

Batch-generate web guide docs and i18n translations

Refactor pages, components, or queries for consistency

Prepare a new release with build, verify, changelog, and git tag

Review changed code against AWSops project conventions

Synchronize project documentation with current code state

Full harness evaluation — multi-agent deep analysis across 12 dimensions (safety, completeness, design quality) with parallel evaluators and synthesized report. Takes 5-10 minutes. Produces a comprehensive scored report with executive summary and improvement roadmap.

Standard harness evaluation — static analysis, dynamic testing, and checklist scoring in 2-3 minutes. Produces a detailed report with findings and improvement roadmap.

Compare harness evaluation history — shows score trends, per-tier deltas, diminishing returns detection, and next grade projection.

Quick harness evaluation — checklist-based scoring in ~30 seconds. Runs deterministic checks against the target project and produces a score, grade, and improvement suggestions.

Manage AWS AI/ML services (Bedrock, SageMaker, Kendra, Q Business) without MCP. Use when the user asks about foundation models, knowledge bases, agents, embeddings, SageMaker endpoints, Kendra search, or Amazon Q.

Troubleshoot and analyze AWS networking (VPC, Transit Gateway, Cloud WAN, Network Firewall, VPN) without MCP. Use when the user asks about network connectivity, VPC configuration, routing, flow logs, security groups, NACLs, transit gateway, cloud wan, or VPN.