Investor AI

The project

Investor AI is an AI-powered stock analysis and prediction platform. My role focused on DevOps: application infrastructure, data processing environments and delivery pipelines.

Infrastructure and data pipelines

• Designed containerized deployment for 12 services, supporting daily financial-data processing through 13 Airflow DAGs.
• Created and maintained Docker images and Compose configurations for consistent environments.
• Prepared GPU- and Spark-enabled containers for reproducible LSTM training and inference, including CUDA configuration and tuning.

Delivery and operations

• Built CI/CD pipelines covering build, test and deployment for backend and frontend, with rollback support.
• Implemented centralized logging, alerting and service-level objectives to monitor reliability.
• Worked with PostgreSQL, Redis, MinIO and the supporting data infrastructure.

What this work brings together

Application delivery, data orchestration and reproducible model environments. The project demonstrates how software and infrastructure have to fit together for a larger system to operate consistently.