In your first 90 days, you’ll complete a landscape assessment of Maple’s current AI features, evaluation coverage, cost profile, and governance gaps. You’ll publish an initial AI standards and best-practices guide covering prompt and agent design, evaluation expectations, safety and PHI handling, and cost hygiene — and stand up an initial evaluation strategy and reference harness that at least one product squad has adopted for an in-flight feature. By month three, you’ll have agreed on the ownership interface with Platform Engineering and built the working relationships across Product, Security, Privacy, and Clinical you’ll rely on going forward.
Over your first 12–18 months, you’ll roll out a company-wide agent execution architecture adopted by the majority of squads shipping AI features, stand up production-grade monitoring for quality, safety, cost, and latency with clear SLOs, and deliver measurable AI cost optimization against your v1 baseline. Every customer-facing AI feature will be covered by an approved evaluation harness and monitoring dashboard, and squads will be independently shipping AI features to Maple’s standards