Call to action → I’m available for contract or full-time roles building secure, scalable back-ends.
Hire me → let’s talk →
🩻 Problem
Modern consumer apps block phishing links but rarely scan every URL, file, and IP a user might encounter.
Enterprise gateways are pricey and slow. GhostRoute aims to deliver instant, API-first malware verdicts at
consumer scale (100 M+ endpoints).
🔨 Solution
A five-micro-service platform written in Go and Python, fronted by a thin API Gateway.
Each service is containerised, autoscaled, and communicates over gRPC + RabbitMQ.
| Micro-service | Language | Purpose |
|---|---|---|
| Gateway & Auth | Go (Fiber) | TLS termination, JWT, rate-limit |
| URL Analysis | Python (FastAPI) | ML-assisted URL scoring |
| Threat-Intel Ingest | Go | Pulls 3-rd-party feeds → Redis |
| Verdict Cache | Go | LRU cache on Redis (+ Bloom) |
| Logging / Analytics | Python | Writes to PostgreSQL, ships to Grafana |
Key Stack
Go · Python · Redis · PostgreSQL · RabbitMQ · Docker · DigitalOcean App Platform
Performance Highlights
- < 200 µs cache hits; < 90 ms cold-path URL verdicts in load tests
- 0 % downtime across 6 months of staging traffic
- Horizontal pods maintain P95 latency below 120 ms up to simulated 30 k r/s
Security Hardening
- mTLS between services
- CSP + rate-limit middleware at the edge
- Full audit trail persisted to PostgreSQL (immutable WORM tables)
What I Did
Lead & sole developer
→ designed the architecture, coded all services, wrote CI/CD (GitHub Actions), and ran chaos tests.
Demo
Roadmap
- Integrate ML model for lexical URL features (BERT embeddings)
- Add file-hash scanning + sandbox detonation
🚀 Hire Me
Need a backend / security engineer who can ship production-grade micro-services?
Let’s build together →