# Braintrust Brand Voice

## Communication Style
*   **Tone:** Authoritative, technical, and pragmatic. Braintrust speaks as a peer to engineers and AI product leaders. The tone is confident and direct, avoiding marketing fluff in favor of performance-driven facts.
*   **Personality:** The "Expert Engineer." It is helpful, methodical, and focused on solving complex problems. It conveys a sense of reliability and architectural maturity.
*   **Vocabulary:** Precision-oriented. Uses industry-standard terminology (e.g., *observability, traces, latency, regression, inference, LLM, eval, span*). It favors active verbs and clear, concise explanations of complex systems.

## Content Patterns
*   **Themes:** Reliability, scale, observability, and the "feedback loop" of AI development. Content consistently focuses on the transition from "experimental" to "production-ready."
*   **Structural Approach:** 
    *   **Problem-Solution:** Identifies a specific pain point (e.g., "AI fails differently," "drift and regression") and positions Braintrust as the necessary infrastructure to solve it.
    *   **Educational/Technical:** Blog posts are long-form, deep dives (10–30 minutes) that function as engineering guides. They emphasize "how-to" mechanics, benchmarking, and architectural design.
*   **Call-to-Action (CTA) Style:** Direct and utility-focused. CTAs are almost always verbs that imply immediate action: *Start building, Log your first trace, Run your first eval, Discover patterns.*

## Audience Interaction
*   **Relationship:** The brand treats the audience as a technical partner. It assumes a high level of competence and does not "dumb down" the content.
*   **Formality:** Professional but accessible. It avoids stiff corporate jargon, preferring the concise, dry wit of the developer community.
*   **Engagement:** The focus is on enablement. The brand interacts by providing tools, documentation, and proof-of-concept benchmarks that allow the user to see the value immediately.

## Guidelines & Examples

### Do's and Don'ts
*   **DO** prioritize technical accuracy and performance metrics.
*   **DO** use clear, concise, and punchy sentences when describing benefits.
*   **DO** link technical features directly to business outcomes (e.g., "Catch issues early").
*   **DON'T** use superlative-heavy marketing language (e.g., "The best platform ever"). Instead, show the data (e.g., "Faster full-text search").
*   **DON'T** obscure technical complexity; acknowledge it and provide the architecture to manage it.

### On-Brand Phrases
*   "Ship quality AI at scale."
*   "AI fails differently than normal software."
*   "Define what good looks like before you ship."
*   "Works with any stack you're already using. No framework lock-in."
*   "From engineering to product, in one platform."

### Content Types
*   **Technical Deep-Dives:** Articles analyzing model comparisons or infrastructure architecture (e.g., *How Brainstore works*).
*   **Practical Guides:** Step-by-step tutorials on using the CLI, SDKs, or specific features like "Topics."
*   **Customer Proof:** Short, impactful testimonials that highlight specific metrics (e.g., "<24hrs to deploy a new frontier model").
*   **Documentation-First:** Marketing copy that mirrors the brevity and clarity of high-quality technical documentation.