# Guides

Short engineering notes on how Arviser scopes LLM quality, AI security, 24-hour PoCs, 40-hour prototypes, and production gates. Orientation, not legal advice.

## 0. AI Security & RAG

| Guide | Focus |
| --- | --- |
| [AI Security Review vs LLM Quality Audit](/ai-security/vs-quality-audit) | Abuse vs performance — when to book which |
| [AI Security Review vs classic pentest](/ai-security/vs-pentest) | AI cybersecurity of LLM apps vs infra/web pentest |
| [AI security risks in plain language](/guides/ai-security-risks) | Business-readable risks R-01–R-27 and questions for your team |
| [OWASP LLM Top 10 2026 — what we check](/guides/owasp-llm-top-10-2026) | Category mapping for scoped reviews (not certification) |
| [Indirect prompt injection](/guides/indirect-prompt-injection) | Instructions via retrieved docs, logs, or tools |
| [Lethal trifecta](/guides/lethal-trifecta) | Untrusted input + privilege + outbound channel |
| [Cross-tenant retrieval](/guides/cross-tenant-retrieval) | When retrieval crosses role or tenant boundaries |
| [AI Security Review](/ai-security) | Product page · RAG track |

## 1. Senior AI Sprints, 24h PoC & Risk-Sharing

| Guide | Focus |
| --- | --- |
| [AI PoC (24h) vs Prototype (40h) vs Production Pilot](/guides/poc-vs-prototype-vs-pilot) | When to book 24h (€2,400) vs 40h (€3,900) |
| [50% risk-shared PoC and sanitization protocol](/guides/risk-shared-poc-and-ip-protocol) | Narrow pipeline pass marks, 50% Kill waiver, IP & synthetic data rules |
| [What a 24-hour AI PoC must prove](/guides/what-an-ai-poc-must-prove) | Four numbers before you build: accuracy, refusal, p95 latency, unit cost |
| [Why 10+ year senior engineers with AI beat vibe-coding](/guides/senior-ai-engineering-vs-vibe-coding) | Senior + AI sprint economics vs fragile demos |

## 2. Evaluation Baselines, Golden Sets & RAG

| Guide | Focus |
| --- | --- |
| [Golden set eval baseline](/guides/golden-set) | What most pilots skip and how to bootstrap 30–100 cases |
| [RAG evaluation](/guides/rag-evaluation) | Gate retrieval and generation separately |
| [Refusal both ways](/guides/refusal-both-ways) | Measure under-refusal and over-refusal in the same suite |
| [Shared LLM quality metrics](/guides/shared-metrics) | Product and risk scoreboard |
| [Cost and latency gates](/guides/cost-latency-gates) | Budget per successful task and p95 latency next to quality |
| [How we score](/guides/how-we-score) | 1–5 scoring rubric across six audit areas |
| [Glossary](/guides/glossary) | Terms we use in audits and sprints |

## 3. Production Gates, Sampling & Package Selection

| Guide | Focus |
| --- | --- |
| [Why a working demo is not a production gate](/guides/when-demo-works-production-fails) | What breaks between demo and real traffic |
| [Readiness vs Audit](/guides/readiness-vs-audit) | Compare 1-week Readiness (€2,900) and 2-week Audit (€8,000) |
| [Two-week audit timeline](/guides/two-week-audit) | Week 1 and week 2 deliverables |
| [Go, hold, or no-go](/guides/go-hold-no-go) | Explicit decision labels tied to pass marks |
| [CI gates for LLM quality](/guides/ci-gates-llm) | Regression harness before scale |
| [Production sampling](/guides/production-sampling) | Async live traffic sampling to catch drift |
| [Kill switch before scale](/guides/kill-switch) | Owner, thresholds, action, and logs |
| [Independent of the build vendor](/guides/independent-of-build-vendor) | Why second-opinion scoring matters |
| [Hiring AI-QA takes months](/guides/hiring-ai-qa) | Sprints, audits, or dedicated specialists while you hire |

## 4. EU AI Act Technical Evidence (Add-on Orientation)

| Guide | Focus |
| --- | --- |
| [EU AI Act dates](/guides/act50-dates) | Orientation table of dates we track for counsel |
| [Art. 50 marking window](/guides/art50-marking-dec-2026) | Dec 2026 transparency and marking orientation |
| [EU AI Act evidence is not certification](/guides/act-evidence-not-certification) | What technical evidence can and cannot claim |

RSS: [/guides/rss.xml](/guides/rss.xml)

**Next:** [Request written scope or question](/#scope) · [AI Sprints](/ai-engineering) · [Startups](/startups) · [Sample deliverables](/sample) · [Checklist](/checklist/view) · [Open proof](/#open-proof)
