← Back to Home design.viswanext.com
Data & AI Architecture Layer

AI & GenAI Architecture

Patterns for putting large language models into production safely — retrieval, orchestration, evaluation, and the guardrails that keep GenAI systems governable at enterprise scale.

Why GenAI needs its own architecture layer

Non-deterministic systems need deterministic guardrails

A GenAI feature is a probabilistic component embedded inside a deterministic enterprise system. Treating it like a normal microservice — no evaluation harness, no cost ceiling, no data-boundary review — is how pilots fail to reach production. The patterns below are the checklist we use before anything ships.

  • Every model call has a token/cost budget and a timeout fallback.
  • Every prompt path is versioned and evaluated like code, not tuned in production.
  • Every retrieval source is access-scoped to the requesting user, not the app.
  • Every output that reaches a customer has a defined human-escalation path.
Core patterns

Reference architectures

🔎
Retrieval

RAG & Hybrid Search

Vector + keyword retrieval, chunking strategy, re-ranking, and citation-grounded generation to reduce hallucination.

🚪
Platform

LLM Gateway

A single ingress for model calls: routing across providers, PII redaction, rate limiting, and audit logging.

🧪
Quality

Evaluation Harness

Offline and online evals, golden datasets, and regression gates so prompt or model changes can't silently degrade quality.

🎛️
Tuning

Fine-Tuning vs. Prompting

Decision tree for when RAG, prompt engineering, LoRA fine-tuning, or full fine-tuning is the right tool for the job.

🧯
Safety

Guardrails & Content Policy

Input/output filtering, jailbreak resistance, and policy enforcement aligned to responsible-AI commitments.

📊
Operations

Cost & Observability

Token-level tracing, latency budgets, and unit-economics dashboards so GenAI spend is a managed line item.

Related standards

What this layer is measured against

NIST AI RMF 1.0 ISO/IEC 42001 EU AI Act OWASP Top 10 for LLM Apps Model Context Protocol NVIDIA NIM
“A model in a notebook is a demo. A model with an evaluation harness, a cost ceiling, and an escalation path is a system.”
— Viswa