🔬
本质：AI 安全公司，不只是 AI 公司
Essence: AI Safety Company, Not Just AI
创始人 Dario/Daniela Amodei 等均来自 OpenAI，因安全理念分歧出走。Safeguards 部门独立存在，招聘生物安全、CBRN 威胁、影响力操控等专职研究员，这在同类公司中罕见。使命优先于商业。
Founders left OpenAI over safety disagreements. Safeguards is a standalone department with dedicated CBRN, bio-safety, and influence-ops researchers — unique in the industry. Mission before money.
💰
战略重心：全面商业化
Strategic Pivot: Full Commercialization
销售部门 150 个岗位，是最大招聘品类，远超研究岗（68个）。遍布 APAC、EMEA、北美，覆盖企业、政府、医疗、金融。已从研究实验室转向全面 B2B 商业化。
Sales has 150 openings — the largest dept, far exceeding Research (68). APAC/EMEA/NA across enterprise, gov, health, finance. Clearly shifted from research lab to full B2B commercialization.
🧠
人才哲学：背景多元，能力为王
Talent: Diverse Backgrounds, Ability First
50% 技术人员没有 ML 经验就加入。50% 有 PhD，但很多没有大学学历。不看标签，看独立研究、开源贡献、实际解决问题的能力。
50% of tech staff had no ML experience when joining. 50% have PhDs, many lack college degrees. No credentialism — they want independent research, open-source, actual problem-solving.
🌍
地理扩张：湾区→全球
Geo Expansion: Bay Area → Global
研究岗集中旧金山，但销售和运营快速国际化：东京、首尔、悉尼、班加罗尔（APAC）；伦敦、都柏林、慕尼黑、巴黎（EMEA）。典型 US SaaS 全球化路径。
Research in SF, but sales/ops are rapidly globalizing: Tokyo, Seoul, Sydney, Bangalore (APAC); London, Dublin, Munich, Paris (EMEA). Classic US SaaS global playbook.
🏗️
组织三大支柱
Three Organizational Pillars
① 研究层（对齐/解释性/RL）→ 能力源头
② 工程层（基础设施/产品工程）→ 工程化
③ 商业层（销售/市场/法务）→ 货币化
商业必须服从安全判断。
① Research (alignment/interpretability/RL) → source
② Engineering (infra/product) → engineering
③ Commercial (sales/marketing/legal) → monetization
Commercial defers to safety judgment.
⚔️
竞争：正面对抗 OpenAI + Google
Competition: Head-to-Head vs OpenAI + Google
差异化是「安全可信赖」，而非功能领先。企业端主打合规、可解释性、数据安全。在垂直行业（金融、生命科学、政府）重点突破。
Differentiation is "safe and trustworthy", not features. Enterprise pitch: compliance, interpretability, data security. Breaking through in verticals (finance, life sciences, gov).