02 · Techniques That Raise Reliability

Technique 1

Few-shot: show, don’t describe

Zero-shot = no examples. One-shot = one worked pair. Few-shot = several. Models pattern-match on examples far more reliably than they follow prose descriptions — the effect that made GPT-3 famous (Brown et al., 2020).

Best for: locking output format, label strings, and edge-case behavior. Three labeled tickets lock BILLING / TECHNICAL / SALES exactly, so downstream code can switch on them.

ticket: "I was charged twice" → {"category":"BILLING"} ticket: "App crashes on login" → {"category":"TECHNICAL"} ticket: "Can we add 50 more seats?" → {"category":"SALES"} ticket: "Password reset email never arrives" → ?

Technique 2

Chain-of-thought: reason before answering

What it is

Ask the model to work step by step before giving the final answer — big gains on multi-step math, logic, and policy questions (Wei et al., 2022). The lazy version works too: just add “Let’s think step by step” (zero-shot CoT, Kojima et al., 2022) — the cheapest first thing to try on a reasoning miss.

When it helps vs hurts

Helps	3-rule discount eligibility · multi-step calculations · ambiguous policy calls
Overkill	Extract a phone number · yes/no lookups — you pay tokens + latency for nothing

For automation: let it reason, then demand a delimited final answer your code can parse — reasoning for the model, structure for the machine.

Technique 3

Structured JSON output: the coupling point

Free text cannot pipe into a database. {"field": value} can. Forcing a fixed, machine-readable shape is what turns “a model that answers” into “a component in a system” — the flow is free text → JSON → database / next agent.

Caution: validate. Well-formed JSON can still contain wrong values — schema conformance is not accuracy (the hands-on tests both).

SCHEMA {"vendor": str, "invoice_no": str, "total_usd": number|null, "due_date": "YYYY-MM-DD"|null, "currency": str} messy input: "Pls remit $1,284.50 to DataCorp by Oct 1 — inv #DC-2291, thanks!!" output: {"vendor":"DataCorp","invoice_no":"DC-2291", "total_usd":1284.50,"due_date":"2026-10-01", "currency":"USD"}

The dial

Temperature: drag it and watch

Temperature controls output randomness. Same prompt — “Name our new project-management app” — different dial positions. Drag it.

0 · deterministic0.5 · balanced1.0 · creative

Run 1

Run 2 — same prompt, run again

Rule of thumb: extraction, JSON, routing, anything automated → 0–0.3. Brainstorming, naming, drafts for humans to pick from → 0.7–1.0. If you’re going to automate it, turn the temperature down.

Synthesis

The reliability ladder

Climb in this order — each rung adds reliability at rising cost. Fine-tuning is the top rung for a reason: try everything below it first.

Clear instructionSay precisely what you want, and what to do when unsurefree · seconds

System promptRole, scope, rules, refusals — the policy guardrailfree · minutes

Few-shot examplesLock format and labels by showing worked pairsfree · minutes

Chain-of-thoughtBuy accuracy on multi-step reasoning with tokenstokens + latency

Structured outputFixed JSON so machines can consume the answerfree · minutes

Tuned temperatureDown for automation, up for ideationfree · seconds

Curated contextThe right facts in the window — next pageengineering time

Fine-tuningTrain the weights — last resort, needs data + upkeep$$ · weeks

Techniques that raise reliability.

Few-shot: show, don’t describe

Chain-of-thought: reason before answering

Structured JSON output: the coupling point

Temperature: drag it and watch

Which technique fixes it?

The reliability ladder