Projects
Projects and production case studies spanning AI-assisted operations, enterprise retrieval, cloud platforms, and reliable delivery.
Featured Projects
RCA Platform
Autonomous AI SRE & Microservices Platform
An autonomous AI SRE platform focused on root-cause analysis across microservices, connecting incident context with production troubleshooting.
- Root-cause investigation across service boundaries.
- AI-assisted context for SRE workflows.
SAP Schema Intelligence
LLMOps / RAG Platform for SAP Metadata
A retrieval-augmented platform for exploring SAP database tables and column information through natural-language questions.
- Natural-language discovery of SAP table and column metadata.
- Schema context for developers, analysts, and integration work.
Production DevOps Case Studies
Elastic GPU Training Platform
GitOps-managed MLOps on Google Cloud
Moved YOLO training from a single on-premises GPU server to a GitOps-managed GKE platform, using queued Spot GPU jobs and scale-to-zero capacity.
- Reduced dataset-to-model time from six hours to two.
- Scheduled training with Argo Workflows and Kueue on NVIDIA L4 Spot pools.
Factory Edge Release Platform
Blue/green delivery for visual inspection
Replaced manual SSH/tmux releases with automated delivery across 50–70 factory inspection stations, preserving live Socket.IO sessions during deployments.
- GitHub Actions self-hosted runners and Docker Compose for repeatable releases.
- NGINX blue/green switching with WebSocket-aware sticky routing.
Academic Inspiration
An attributed example from Concordia's published Electrical and Computer Engineering capstone archive.
IoT Monitoring for First Responders
Concordia ECE · Capstone reference (2022)
A Concordia ECE capstone example using wearable sensors and wireless telemetry to monitor first responders. Included as academic inspiration.
- Connected sensing, wireless communication, and a monitoring backend.
- Original project: Concordia ECE Group 2022-13, supervised by R. Coutinho.