Prompt Engineering

NCP-GENL · 26 questions

• A permitting office wants the model to check zoning district, setbacks, then fee class before answering. Which prompt pattern best structures that multi-step reasoning?
• Public works has only a few dozen labeled culvert-inspection notes and wants to adapt a model without full fine-tuning. Which approach fits?
• A city clerk needs meeting-minutes Q&A answers that cite ordinance section IDs inside a fixed output schema. What should the prompt template enforce?
• Housing caseworkers need a reusable prompt wrapper for eviction-resource chat that sets role, tone, and prohibited-advice boundaries. What should that template include?
• Fire inspection prompts should separate checklist reasoning from the final pass/fail narrative. Which technique supports that separation?
• A school nutrition bot must follow USDA meal pattern rules rather than generic diet tips. How should prompts be adapted?
• Transit paratransit booking prompts must ensure policy checks happen before ride-time suggestions. What ordering should the prompt enforce?
• County open-data portal chat should prefer CSV column glossary terms over invented field names. What system-prompt guidance helps?
• A brand-new stormwater fee FAQ goes live before any curated answer examples exist. Which prompting regime is appropriate to start?
• Clerk staff want to stabilize tone on ballot-explanation answers using a single gold sample. Which technique fits?
• Library reference chat must refuse medical diagnosis while still pointing residents to vetted links. How can few-shot prompting help?
• After residents confuse trash versus recycling rules, public works wants the model to distinguish correct sortation answers. What few-shot strategy helps?
• Emergency SMS rewrite starts with no approved alert phrasings, then later gains three ratified gold alerts. How should prompting evolve?
• A parks botanic-garden chatbot keeps mangling Latin plant names in resident answers. Staff have one gold reply that shows the required Genus species formatting. Which prompting change best fixes this narrow formatting failure?
• A county assessor wants appeal-letter drafts that sound locally empathetic rather than like a generic national template. Which prompting approach best improves adaptability for this civic tone?
• City IT wants to continue pretraining a decoder LLM on municipal code text so the model learns local ordinance phrasing before any task-specific prompts. Which training objective should they use?
• A transit agency is further adapting an open decoder-only checkpoint on rider-complaint narratives. Engineers debate masked language modeling versus causal language modeling. Which choice is appropriate?
• Public-health continued pretraining shuffles clinic-note documents across the batch but must still train a decoder with causal language modeling. How should token order be handled?
• A courts IT team wants a decoder LLM to absorb public opinion phrasing before any labeled classification work. Which approach best matches that goal?
• A permitting desk requires every model status line to be exactly one of Approved, Needs revision, or Denied. Which wrapping technique best enforces that consistency?
• A 311 assistant sometimes invents service-request types that are not in the city’s official taxonomy. What should the LLM wrapping module do before showing a suggestion to residents?
• A housing-authority chatbot sometimes drops rent-limit dollar amounts into casual prose where staff cannot parse them reliably. Which wrapper design best improves consistency?
• A library hours bot invents branch codes that do not exist in the facilities table. Which technique best blocks those invented IDs?
• An emergency-alert rewriter sometimes inserts phone numbers that are not on the approved contact list. What should the wrapping module do before send?
• A tax FAQ wrapper must not show ordinance answers that lack retrieved citation spans. Which behavior should the module implement?
• A water-quality chatbot must name contaminants only from EPA or state code lists. Which decoding approach best enforces that regulated vocabulary?