Prompt Cards

Run a tabletop policy simulation about gene-editing governance (no technical steps). Deliver: stakeholders, scenarios, decision trees, outcomes, comms plan, ethics/legal brief, KPIs (transparency, risk reduction). Output: facilitation guide, scoring sheet, debrief template.

Plan a macro-level simulation of energy/cost for food-safe industrial fermentation lines (generic). Deliver: process blocks, inputs (power tariffs, throughput), KPIs (kWh/kg, cost/unit), sensitivities, Pareto. No recipes, organisms, or lab steps. Output: model spec, data templates, dashboards, 2-week pilot plan.

Create an educational ODE sandbox for gene regulation toggles using dimensionless forms. Deliver: equations, parameter sweep plan, stability analysis, plots, teaching notes. No plasmids, sequences, or lab parameters. Output: notebook outline, sliders spec, learning outcomes, assessment rubric.

Design a purely educational docking simulation plan with public PDB examples only. Deliver: objective, dataset list, feature extraction, baseline (shape complementarity), KPIs (rank@k), error analysis, visualization plan. Prohibit novel sequence design or wet-lab suggestions. Output: pipeline diagram, compute budget, 10-day plan, reproducibility checklist.

Plan a harmless SIR/SEIR toy simulation for campus flu dynamics using public, anonymized stats. Deliver: scope, compartments, equations, parameters to estimate, scenarios (mask uptake, class density), KPIs (R_t, peak load), charts, uncertainty. No pathogen design, no lab steps. Output: code outline, CSV schema, run matrix, ethics/comms page for public understanding.

Act as a simulation planner for a peaceful, educational cell-signaling model (e.g., MAPK cascade). Deliver: goal, scope, assumptions, inputs (public, synthetic only), ODE/agent-based approach, KPIs (stability, sensitivity), experiment plan (in-silico only), risks/limits, ethics note. Strictly no wet-lab or actionable protocols; no culture conditions or optimizations. Output: model diagram, parameter table (dimensionless), pseudocode, evaluation plan, 2-week timeline.

Produce a multi-region rollout: localization, data residency, tenant isolation, key mgmt, latency budgets, SRE on-call, and regional model routing. Include training paths, comms plan, and a lighthouse-customer program with ROI commitments.

Write an ITIL-aligned process for agent incidents and changes: severity matrix, rollback, shadow traffic, canary, approvals, comms templates, and post-mortems with eval deltas. Include regulator-ready evidence checklist.

Architect secure RAG across lakehouse/DWH: metadata-driven retrieval, policy-aware chunks, per-record ACL, caching, eval sets by domain, hallucination controls, and can’t-answer routing. Deliver reference pipelines and SLOs.

Create a decision brief comparing vendor platforms vs internal build across control, speed, TCO, compliance, talent, and lock-in. Provide a hybrid option. Output: scorecard, phased migration path, and exit strategy clauses.

Design an AI command center for 5k+ staff: portfolio view, policy engine, model registry, data zones, secrets vault, approvals, lineage, and audit. Include risk taxonomy (legal, security, brand), RACI, escalation flow, and board-level KPI set.

Build a FinOps plan for AI: unit economics per task, cost alerts, usage quotas, fallback models, and A/B testing harness. Output dashboards (tasks/hour, cost/task, quality), run-book for anomalies, and a monthly optimization ritual.