Prompt Cards

Write a procurement playbook: evaluation criteria, red flags, security review, data terms, SLAs, pricing traps, and exit clauses. Include a scoring matrix and interview questions for vendors.

Create an AI governance program aligned to common frameworks: policy stack, model inventory, risk register mapping, approvals, audits, and incident management. Provide templates and RACI.

Draft a 2026 cost engineering playbook: caching, quantization, distillation, batching, and SLA tiers. Provide a KPI dashboard linking cost per task to business outcomes.

Propose a model routing architecture (‘model mesh’): task-based routing, fallback policies, confidence thresholds, and cost controls. Include an observability plan for routing quality.

Create an enterprise unstructured data strategy: ingestion, classification, redaction, indexing, provenance, and access control. End with a ‘document-to-decision’ workflow and KPIs.

Design a hybrid retrieval strategy combining vector search, lexical search, and a knowledge graph layer (‘Graph-RAG’). Specify when to use which, how to evaluate, and how to keep citations reliable.

Create an eval-first LLMOps design: golden sets, adversarial tests, continuous regression, cost/latency tracking, and release gates. Include a ‘model change control’ policy.

Build a P&L-style ROI model for deploying agents as digital workers: labor displacement vs augmentation, quality risk, error cost, cycle time reduction, and governance overhead. Provide a template with assumptions.

Create a physical AI roadmap for 2026: use cases, safety constraints, simulation strategy, edge compute needs, and incident handling. Include a phased pilot plan with success criteria.

Design a preemptive cybersecurity approach: AI-driven exposure management, automated patch prioritization, agent-assisted triage, and ‘human-in-the-loop’ escalation. Include KPIs that reduce mean time to contain.

Write requirements for an AI security platform: prompt/tool injection defenses, model access control, secrets handling, audit trails, red-team workflows, and runtime policy enforcement for agents.

Define a digital provenance program: signing, watermarking policies, asset lineage, approval workflows, and incident response for deepfake risk. Provide a policy template and metric dashboard.