DEV Community: Jackob Sumner-Knapp

Writing Prompts That Survive Contact With Users: a practical guide

Jackob Sumner-Knapp — Wed, 26 Aug 2026 09:44:03 +0000

Writing Prompts That Survive Contact With Users

A prompt is a program written in natural language, and the first version never survives real input. This course covers the practical prompt engineering for production — versioning, adversarial inputs, the few-shot set that earns its place, and the eval that tells you when a change regressed — that distinguishes a demo from a feature you can ship and keep.

Why this gap matters now

Demand signals we detected:

prompt-fragility complaints up 210%
jailbreak/regression incidents recurring
sparse content on prompt engineering for production vs demos

What you'll learn

Treat the prompt as versioned code — An unversioned prompt means every change is an unobserved regression, and when the feature breaks next week nobody can find what changed.
Few-shot earns its place, it doesn't pad it — Few-shot examples are the most expensive tokens in your prompt and the most often cargo-culted.
Plan for the input you didn't expect — Real users send inputs the demo never imagined: empty strings, pasted documents, other-language text, and deliberate adversarial phrasings.
The regression eval that guards every change — Every prompt change — a phrasing tweak, a new example, a model swap — can regress a class of inputs that worked before.

Who it's for

Engineers and PMs shipping LLM features to real users, 1–5 yrs.

Start learning

Read the full micro-course and earn a certificate of completion: Writing Prompts That Survive Contact With Users.

Want the full structured micro-course with a certificate of completion? Open it on SkillForge — $25, instant access.

Cost-Aware Prompting for Production LLM Apps: a practical guide

Jackob Sumner-Knapp — Wed, 26 Aug 2026 09:17:52 +0000

Cost-Aware Prompting for Production LLM Apps

An LLM feature that works in demo blows the bill in production because nobody priced the prompt. This course covers the token-level engineering that keeps a feature profitable: where tokens actually go, prompt shaping for cost, caching the repeated context, and the model-routing decision that returns most of the saving.

Why this gap matters now

Demand signals we detected:

token-bill threads up 200%
production cost-vs-quality teams forming
scarce content on token-level engineering tradeoffs

What you'll learn

Where the tokens actually go — Most prompt cost is not the user's question; it is the context you re-send every turn — system prompt, retrieved passages, examples — billed in full on every call.
Prompt shaping for cost — A verbose prompt is not just slow, it is expensive, and the model often performs no better for the extra words.
Cache the repeated context — The context you send identically every call is the cheapest thing to cache.
Route by difficulty, not by default — Routing everything to the strongest model is the most expensive default.

Who it's for

Engineers running LLM features against a token budget, 2–6 yrs.

Start learning

Read the full micro-course and earn a certificate of completion: Cost-Aware Prompting for Production LLM Apps.

Test publish (please ignore)

Jackob Sumner-Knapp — Wed, 26 Aug 2026 09:16:46 +0000

test publish