Frequently Asked Questions

GoNoGo ensures determinism through snapshot-based grounding and structured evaluation criteria. Each test run is compared against a canonical snapshot, removing variability from LLM responses. The evaluation framework uses predefined rubrics that produce consistent pass/fail decisions regardless of when or how many times you run the test.

Persona packs define different user archetypes (e.g., power users, beginners, skeptics) with specific behavioral traits and expectations. When you run a test with a persona pack, GoNoGo simulates how each persona type would interact with your AI and evaluates responses against persona-specific criteria.

Yes! GoNoGo supports custom test type definitions via JSON schemas. You can define your own evaluation criteria, scoring rubrics, and pass/fail thresholds. Custom test types can be registered through the API or the dashboard.

When personas produce conflicting evaluations, GoNoGo provides detailed conflict reports showing exactly where and why disagreements occurred. You'll see side-by-side comparisons, the specific criteria that caused divergence, and confidence scores for each evaluation.

Confidence scores are derived from multiple factors: consistency across repeated evaluations, alignment with grounding snapshots, persona agreement levels, and evaluation criteria clarity. Scores range from 0-100%.

Absolutely. GoNoGo supports multiple export formats including JSON (for programmatic access), PDF (for stakeholder reports), and CSV (for spreadsheet analysis). Reports include full evaluation details, confidence metrics, and persona breakdowns.

Yes, an account is required to access the GoNoGo dashboard and run tests. Each test is priced transparently based on complexity, and you see the exact cost before running. Sign up takes less than a minute.

GoNoGo uses a pay-as-you-go model with transparent per-test pricing. Single tests start at $9, or save 45% with a 10-test bundle at $49. Costs vary based on test complexity, persona count, and LLM models used.

Yes, you can cancel your subscription at any time with no penalties or hidden fees. Your access continues until the end of your current billing period.

GoNoGo integrates with major LLM providers including OpenAI (GPT-4, GPT-4 Turbo), Anthropic (Claude 3 Opus, Sonnet, Haiku), Google (Gemini Pro, Ultra), and open-source models via API.

Yes, GoNoGo provides a comprehensive REST API that integrates with any CI/CD pipeline. We offer official GitHub Actions, GitLab CI templates, and Jenkins plugins.

Security is paramount at GoNoGo. All data is encrypted in transit (TLS 1.3) and at rest (AES-256). We never use your test data to train models. Enterprise plans include SOC 2 compliance and data residency options.