I am a Forward Deployed AI Engineer and Cloud Systems Architect, and the technical founder of GG Loop LLC. I specialize in building real-time event telemetry streaming pipelines, agentic LLM architectures (MCP), and enterprise data automation.
I bridge deep distributed cloud infrastructure with high-stakes financial and game-system execution.
- β‘ Real-Time Telemetry & Event Streaming: Architected sub-50ms event ingestion on AWS using Kinesis Data Streams, Lambda, and WebSockets with zero dropped events.
- π€ Autonomous Agentic AI (MCP): Production multi-agent orchestration integrating AWS Bedrock (Claude 3.5 Sonnet) and OpenAI APIs via Model Context Protocol (MCP), featuring dynamic schema validation, automated tool calling, and pgvector RAG retrieval.
- π¦ Enterprise Financial & Systems Automation: Former Finance Associate II at NCR Atleos, reconfiguring bank routing parameters for national ATM networks, managing multi-million-dollar cash settlements, and cutting discrepancy investigation cycle times by 40% with Python/SQL reconciliation pipelines.
- π‘οΈ Client Integrity & System Hooks: Developed Electron.js native desktop applications paired with Node.js and C++ user-space (Ring 3) process watchdogs, memory buffer management, and cryptographic HMAC-SHA256 verification.
- βοΈ Cloud & Container Scaling: Zero-downtime containerized deployments across AWS App Runner, ECS Fargate, RDS PostgreSQL, and CloudWatch with Docker and GitHub Actions CI/CD.
- Problem: Live-service games and esports tournaments require high-throughput match state auditing without invasive kernel drivers or expensive bloated telemetry suites.
- Architecture: Ingestion through AWS Kinesis Data Streams and AWS Lambda fanout to WebSocket state synchronization, backed by RDS PostgreSQL and pgvector for match embeddings.
- Result: Sub-50ms ingestion latency, 100% idempotent Stripe Connect payout execution, and auto-scaling serverless infrastructure.
- Scope: Daily settlement reconciliation across national banking and ATM networks.
- Impact: Replaced manual exception tracking spreadsheets with automated Python/SQL anomaly detection pipelines, cutting investigation cycle times by 40% with 99.9%+ balance sheet accuracy.
| Repository | Description | Tech Stack |
|---|---|---|
| π³ stripe-webhook-guard | Production-grade zero-loss Stripe webhook idempotency, cryptographic verification, and replay suppression. | TypeScript, Node.js, Express, Stripe API |
| π‘οΈ anti-cheat-sdk | Lightweight zero-overhead process scanning and active-window focus tracking SDK for Windows and macOS. | C++, Node.js, Rust |
| π¦ freely-reconciliation-engine | Enterprise B2B payment reconciliation & double-entry ledger verification engine. | TypeScript, SQL, PostgreSQL |
| βοΈ aws-infrastructure-playbook | Production AWS patterns: App Runner, ECS Fargate, RDS, Bedrock, and automated FinOps cleanup. | Python, Boto3, AWS CloudFormation |
| π₯οΈ electron-resilience-toolkit | Fault-tolerant desktop process manager with heap monitoring, automated watchdogs, and CI truth gates. | TypeScript, Electron, Node.js |
| π sql-reconciliation-engine | Database schema reconciliation and financial transaction auditing engine inspired by NCR Atleos operations. | SQL, PostgreSQL, Python |
- Forward Deployed AI Engineer / Solutions Architect: AI infrastructure, enterprise model deployment, agentic systems (MCP).
- Distributed Cloud / Systems Engineer: Real-time event streaming, telemetry pipelines, high-throughput microservices.
- Location: Dallas-Fort Worth, TX (Open to Remote, SF, NYC, Tokyo, or Relocation).
π¬ Get in touch: jquindao1@icloud.com β’ LinkedIn β’ Portfolio


