Platform SRE Build-and-Transfer Director

The mandate

The company plans to build SRE in Chennai and reduce dependence on vendors who currently hold production access, incident memory and platform recovery. Hiring roles before defining service ownership risks creating a monitoring team without decision authority. The defined problem is to design a capability that can accept complete services, command incidents and improve reliability while vendors exit safely.

The named deliverable is a Chennai Platform SRE Build-and-Transfer Blueprint and Service Acceptance Kit. It must define service tiers, SRE and product rights, leadership, staffing, on-call, access, observability, error budgets, incident and recovery, vendor knowledge, economics, launch waves, source exit and five executable acceptance scenarios.

Four milestones govern six months: by week three, accept service and dependency evidence; by week ten, deliver capability and leadership architecture; by week eighteen, run acceptance scenarios for two services; and by week twenty-six, submit the blueprint, hiring sequence, vendor-exit plan and investment paper. Each invoice follows milestone acceptance.

The platform officer and launch council will accept the work only when Chennai can deploy, observe, diagnose, recover and govern two services within signed thresholds, vendor access and knowledge have measurable exit conditions, leadership and on-call capacity are viable, product teams accept shared reliability rights, and client teams can apply the kit to a third service unaided.

The client will provide service maps, incidents, access models, contracts, runbooks, observability, staffing, budgets, reliability data and engineers from vendors, source teams and Chennai. The consultant will not recruit staff, operate production, terminate contracts, implement tooling or approve service transfer; gaps remain explicit launch dependencies.

Why this is external work

Vendors describe current operation and have an interest in continued scope, while product teams want reliability without surrendering release discretion. Chennai recruitment is already moving faster than operating design. External SRE GCC expertise is required to create complete service acceptance and an executable vendor exit before roles are filled around a weak charter.

What you will own

• Classify services by customer consequence, persistent state, failure modes, support demand and the safest Chennai acceptance wave.
• Define SRE, product, platform, security and business decision rights for error budgets, reliability investment and release.
• Design leadership, staffing, on-call, escalation, learning reviews and two-deep coverage for every production service tier.
• Map vendor-held credentials, tooling, incident heuristics, supplier relationships and recovery actions into observable exit evidence.
• Build scenarios for failed release, dependency loss, capacity saturation, security containment and cross-region recovery under pressure.
• Model capability economics including duplicate running, attrition, training, observability tools, vendor transition and avoided reliability loss.
• Deliver the operating blueprint, service-acceptance kit, two exercised transfers, hiring sequence and unresolved production-risk register.

Candidate qualifications

• Built an India GCC SRE capability that accepted stateful production services from vendors or overseas platform teams.
• Defined product and SRE authority for error budgets, release risk, capacity and reliability investment rather than support queues.
• Designed incident-command, on-call and service-owner depth before volume hiring and migration waves created fragile coverage.
• Extracted undocumented vendor recovery heuristics through exercised failovers, failed deployments and dependency-loss scenarios.
• Linked reliability investment to customer impact, recovery performance, engineering flow and fully burdened transition economics.
• Delivered a service-acceptance kit that internal leaders reused to approve, defer or reject later production migrations.

Non-negotiables

• Can complete two source-team residencies and all Chennai milestones within six months.
• Will disclose logistics, SRE, cloud, vendor, recruitment and GCC advisory relationships.
• Brings production SRE build-and-transfer into India; monitoring operations design alone is insufficient.
• Accepts client-run service scenarios and vendor-exit evidence as final acceptance conditions.

1. 49 words maximum. Describe a service a new SRE centre should refuse until vendor-held recovery knowledge is exercised.
2. 49 words maximum. Which five scenarios would you include in Chennai’s first service acceptance kit?
3. 49 words maximum. How would you prove an SRE role carries decision authority rather than escalation duty?