Skip to content
All projects

AI Product

Zizta

AI SEO platform

An AI research and content system that drafts, critiques and grades briefs against real SERP data - with evaluation baked into the pipeline.

  • Laravel
  • React
  • OpenAI
  • PostgreSQL
  • Redis
  • Cloudflare
Visit live product

The problem

AI content tools produce confident, generic drafts. Teams could not tell which output was actually good before publishing it.

Business challenge

Make quality measurable: every generated brief needed a score a human could trust and argue with.

Research

Built a labelled set of 600 briefs graded by editors, then tuned an evaluation rubric until the model agreed with editors 90% of the time.

Planning

A multi-stage pipeline - retrieve, draft, critique, revise, score - with each stage cached, observable and independently replaceable.

Architecture

  • Queue-driven generation pipeline with per-stage retries
  • Vector retrieval over crawled SERP and client corpus data
  • Model routing: cheap models for extraction, strong models for critique
  • Full prompt/response logging for evaluation and cost attribution

Database design

PostgreSQL with pgvector for embeddings, plus a run table capturing every stage's input, output, tokens and cost.

API flow

  1. 01User submits a keyword and target audience
  2. 02Retrieval stage assembles SERP context and internal corpus chunks
  3. 03Draft → critique → revise loop runs until the score plateaus
  4. 04Final brief and its score card are returned and stored

Implementation

  • Started with evaluation harness before writing a single prompt
  • Streaming UI so users watch reasoning instead of waiting
  • Per-workspace cost caps to keep unit economics predictable

Key features

  • SERP-grounded content briefs
  • Automated critique with a transparent score card
  • Workspace-level cost and usage controls
  • Streaming generation UI

Performance

  • Model routing cut token spend without changing output scores
  • Cached retrieval reduced repeat-run latency by 3x

Lessons learned

  • Build the evaluation before the feature. It changes every later decision.
  • Cheap models do 70% of the work if you route deliberately.

Taking on 2 new partnerships for Q4

Let's talk about what you're building

Thirty minutes, no pitch deck. Bring the problem - you'll leave with an honest read on scope, risk and the fastest credible path to launch.