18+ years turning ambiguous, high-risk infrastructure into predictable, autoscaled systems — with tight cost controls, strong security baselines, and SLO-driven reliability.
Most infrastructure is built to handle average load. Mine is built for the moments that matter: election nights, IPL finals, breaking news — when failure is not an option.
Designing cloud platforms for media-tech, commerce, and SaaS that handle millions of users without degradation. Multi-cloud architecture, microservices decomposition, and platform engineering that treats infrastructure as a product.
SLO definition, incident management transformation, runbook engineering, and on-call culture design. Turning reactive firefighting into calm, measured operational execution — even at 2M concurrent users.
FinOps-led cost optimization that doesn't trade performance for savings. Rightsizing, autoscaling economics, waste elimination, and building engineering cultures where cost is a first-class signal — not a monthly surprise.
WAF hardening, Zero-Trust implementation, CDN edge strategy, and security postures aligned to real-world threat models. Akamai Traffic Peak readiness and protection at scale — without slowing delivery.
Cache hit ratio at 30%, origins overloaded during bursts, autoscaling reacting too late. Cost rising from fear-based provisioning.
Rebuilt CDN + edge caching from scratch. Cache key normalisation, TTL tuning by content pattern, WAF pre-filtering junk traffic.
Replaced CPU-based scaling with request rate, queue depth, and downstream health signals. Added pre-warming for predictable events.
Golden signals per service, low-noise alerting, SLO-aligned runbooks, and cost as a first-class engineering signal.
The exact 5-layer framework we used to cut infrastructure costs 30% at Network18 — across 25+ high-traffic portals. Written for engineering leaders, not analysts.
How We Cut Infrastructure Costs 30% Without Touching Performance — A 5-Layer FinOps Framework for High-Traffic Platforms.
Traffic in media doesn't grow gradually. It explodes. Here's what changed the game at Network18.
We didn't add servers. We stopped wasting the ones we had. Here's the exact lever.
AI generates infrastructure for the average case. Your production system is never the average case.
"Our AWS bill tripled since we added the AI feature and nobody can explain why."
I help you build the cost governance architecture that makes your infrastructure spend predictable — before the CFO meeting becomes uncomfortable.
"Elections, IPL finals, breaking news — we're never confident we're ready for peak load."
I build the edge strategy, autoscaling playbook, and pre-warming runbooks that turn traffic spikes from emergencies into operational routines.
"We're still running the architecture decisions we made in 2019 on infrastructure that was never designed for this scale."
I lead cloud modernization that stabilizes before it transforms — reducing risk, maintaining availability, and building toward the platform your next phase requires.
If your infrastructure is the thing between your business and its next level — let's talk.