Infrastructure Specifications
Every technical decision we make is documented here. No marketing language. If you need to know whether you can run production systems, conduct live interviews, or push to CI/CD from Ladakh — this page answers it.
01 · Network Architecture
## Network Topology
Internet ──► [BSNL 300M] ──► pfSense Router ──► LAN Switch ──► AP (Wi-Fi 6)
└──► [Jio 200M] ──► (failover, hot standby)
# Failover logic
monitor: ping 8.8.8.8 + 1.1.1.1 every 500ms
threshold: 3 consecutive failures → automatic switch
switch_time: < 4 seconds
02 · Cloud Region Latencies
Measured via ping and mtr. Average of 100 samples, 08:00–20:00 IST. Last updated: May 2026.
| Cloud Region | Ladakh | Goa | SF (ref) |
|---|---|---|---|
| AWS ap-south-1 (Mumbai) | 42ms | 28ms | 210ms |
| GCP asia-south1 (Mumbai) | 45ms | 30ms | 205ms |
| Cloudflare Edge (India) | 18ms | 12ms | 85ms |
| Vercel Edge (Asia) | 35ms | 22ms | 95ms |
| GitHub (Global CDN) | 55ms | 40ms | 30ms |
| npm registry | 60ms | 45ms | 25ms |
✅ = sub-60ms (excellent for live coding, pair programming, video calls). AWS Mumbai is the optimal region for workloads running from India.
03 · Workspace Ergonomics
04 · What You Can Run From Here
✅ Production CI/CD
GitHub Actions, CircleCI, Jenkins — all tested. Push-to-deploy with sub-5s feedback loops.
✅ Live Video Calls
Zoom, Google Meet, Loom screen recording. Dual-ISP ensures zero dropped calls.
✅ Remote Desktop
AWS WorkSpaces, Parsec, RDP — all usable. Sub-50ms to Mumbai region is gaming-grade.
✅ Docker / K8s
Full Docker build speeds on-site. kubectl to cloud clusters without issue.
✅ Database Connections
Supabase, PlanetScale, MongoDB Atlas — all reachable with excellent latency.
⚠️ Latency-Sensitive HFT
If your work requires sub-10ms global latency (trading algos), colocate your compute in Mumbai.
The infrastructure is ready.
Are you?
Applications open for the July 2026 Ladakh cohort.