Rahul Ladumor · AWS and production GenAI
I help SaaS and enterprise teams fix recurring failures, control cloud costs, and get AI features safely into production.
From $2,000 · one bounded system · 10 business days
AWS Professional certified · 5x AWS Community Builder (Serverless)A release has no demonstrated recovery path.
Proposed first step
Test the existing rollback in staging.
01 / Selected work
A simplified view. Full architecture and trade-offs in the case study.
GenAI / AI · 2024
30+ min document hunts → sub-10s answers; hybrid RAG + semantic cache cut LLM spend ~60%.
Reported engagement figures. Measurement context and limitations are in the case study.
Read the engineering decisions02
Migration · 2022
Peak-hour monolith failures + inflated EC2 → 5M+ orders/mo, P99 <200ms, ~30% infra reduction.
Read the case03
Cloud Architecture · 2025
Manual VPC sprawl + quarterly audit pain → standardized IaC, 70% faster setup, findings down 47 → 3.
Read the caseQuotes reflect real engagements. Named references available under NDA for qualified inbound.
02 / Working together
You share the architecture and the problem. I review the evidence and give your engineers a written fix order. Implementation is scoped separately.
Look inside the deliverableStart here / one bounded system
From $2,000
10 business days
Is the AWS bill the main problem?
A focused cost audit: from $1,500, with a plan in 7 business days.
See the focused cost auditWe agree on deliverables, access, timing and fee before work begins.
Move a mature AI prototype into controlled production with narrower tool permissions, stronger retrieval evidence, evals, fallbacks, human approval, and workflow-level cost controls.
Paid engagement · fixed scope
Discuss productionizationStabilize an expensive, fragile, undocumented, or partially abandoned AWS platform without beginning with a risky rebuild.
Paid engagement · scoped phase
See how project rescue worksPrincipal-level AWS and GenAI ownership without waiting for a full-time hire or bringing in a large consultancy.
Monthly retainer · scoped hours
Discuss fractional ownershipFinance sees the number. Engineering sees hundreds of services. Nobody has a safe first move.
I start with cost allocation, idle inventory, traffic shape, and commitment risk before touching architecture.
The feature looks impressive in a meeting and then becomes expensive, slow, or untrusted with real users.
I inspect retrieval, context design, evals, fallbacks, and cost per request before blaming the model.
Every release needs extra human attention. That slows the team and creates fear.
First I check blast radius, deployment path, stateful dependencies, and rollback ownership.
Dashboards are present, but incidents still become log archaeology and Slack guessing.
I connect symptoms to traces, metrics, logs, ownership, and alarms that point to action.
Services and AI tools can reach more than they should, while auditability and human approval remain unclear.
I define account boundaries, least-privilege tool access, approval controls, audit evidence, and IaC ownership.
Growth becomes a threat because failure modes, queues, database limits, and on-call paths are vague.
I model what fails first, then build guardrails around scale, retries, backpressure, and recovery.
03 / Client perspective
₹50L+/year in AWS waste removed for clients (~$60K+/year), from cost and reliability audits.
“Rahul migrated our entire ordering pipeline to serverless and cut our AWS bill by 30% while handling 5M+ orders per month. He doesn't just architect: he ships production infrastructure.”
Quotes reflect real engagements. Named references available under NDA for qualified inbound.
04 / Before we start
Architecture evidence and read-only access where it helps, bounded in writing before anything starts. In 10 business days you get a written risk register, a defensible target architecture, and a prioritized remediation roadmap, plus a 90-minute walkthrough and two weeks of async follow-up questions.
Report first: it keeps my incentives honest. Implementation support and fractional-architect retainers are available afterward, and most clients use their own team with my report as the roadmap.
I specialize in systems that need to survive production: serverless migrations, AWS cost optimization audits (typically 30-40%, up to 55% documented), event-driven architectures on ECS/EKS, GenAI/RAG infrastructure on AWS Bedrock, and multi-account cloud security. My value is highest when you need reliability, security, and cost control at scale.
Typical AWS + LLM engagements land around 30 to 40% when waste is structural: idle compute, wrong DB tier, unbounded model calls, missing lifecycle. On documented work, reductions reach up to 55%. I quote from utilization and billing data, not from a template. ₹50L+/year in AWS waste removed for clients (~$60K+/year), from cost and reliability audits.
A short description is enough to start. We agree on scope, access and fee before the review begins.
Surat, India (IST) · US / EU / APAC remote
Typical reply: one business day.