final year project
Logarda
A cloud monitoring system, observed end-to-end — agents, ingestion, storage, forecasting, LLM inference, and live alerting in one platform.
PythonBashDockerCI/CDAWS ECRAWS EC2PostgreSQLRedis
shipped
- Automated data pipeline ingests compute metrics & logs into PostgreSQL
- Error logs queued in Redis for scalable downstream processing
- Containerized with Docker, deployed to ECR, runs as CRON on self-hosted EC2
in flight
- Data processing pipeline preparing metrics for time-series forecasting
- LLM inference pipeline backed by a RAG workflow over AWS documentation
- API endpoints for LLM inference on error log messages
GolangPostgreSQLRedis
shipped
- Scalable backend supporting concurrent users
- AES-GCM encryption and SHA-256 hashing for data security & integrity
in flight
- Metrics stream worker pushing live updates to the dashboard
- Error message inference worker monitoring the Redis queue
- API endpoints handling client-side requests in real time
development progress
Build status · v0.4
4 shipped · 4 in flight
Completed
- Metrics ingestion● shipped
- Docker deployment● shipped
- EC2 deployment● shipped
- Security implementation● shipped
In Progress
- Forecasting engine◐ building
- LLM inference service◐ building
- Redis workers◐ building
- Real-time dashboard updates◐ building