Remove dating/relationship references from all files

AGENT.md

@@ -49,7 +49,7 @@ This downloads the 1M-persona dataset (~2GB). Only needs to happen once.
 
 **Ask the user**:
 
-1. *"What are you optimizing? (product, resume, pitch, policy, dating profile, or describe your own)"*
+1. *"What are you optimizing? (product, resume, pitch, policy, profile, or describe your own)"*
 2. *"Describe it — or paste/point me to the document. I need what an evaluator would see."*
 3. *"Is there anything an evaluator should NOT see? (internal metrics, private details, etc.)"*
 
@@ -80,7 +80,7 @@ If the user doesn't have a document ready, use the appropriate template from `te
 | Resume | Company type, seniority, technical depth, industry |
 | Pitch | Investment stage, sector focus, check size |
 | Policy | Stakeholder role, income bracket, geography, property ownership |
-| Dating | Age bracket, life stage, education, occupation, geography |
+| Profile | Age bracket, life stage, education, occupation, geography |
 | Custom | Ask the user to name 3-4 dimensions |
 
 ### Build the cohort

README.md

@@ -35,7 +35,7 @@ Anything someone else evaluates.
 | **Pitch** — investor deck | VCs and angels at different stages | Whether the story lands, what questions they'd ask |
 | **Policy** — proposed regulation | Stakeholders by role, income, geography | Who supports it, who opposes, what compromise works |
 | **Content** — blog post, video | Readers at different expertise levels | Whether it hits the right level, what's confusing |
-| **Profile** — dating, professional bio | Population sample by age, education, occupation | How different demographics perceive you |
+| **Profile** — professional bio, personal brand | Population sample by age, education, occupation | How different demographics perceive you |
 
 SGO ships with a 1M-person census-grounded dataset ([Nemotron-Personas-USA](https://huggingface.co/datasets/nvidia/Nemotron-Personas-USA)) with structured demographics (age, sex, education, occupation, marital status, US geography) plus rich narrative fields — professional persona, skills and expertise, career goals, hobbies, cultural background, and personality. The narratives naturally encode things like seniority, industry, technical depth, and decision-making style, even though those aren't separate columns.
 

scripts/generate_cohort.py

@@ -63,7 +63,7 @@ Return JSON:
             "country": "USA",
             "education_level": "<high_school | bachelors | graduate | etc>",
             "occupation": "<specific job title>",
-            "marital_status": "<never_married | married | divorced | widowed | separated>",
+            "marital_status": "<single | married | other>",
             "interests": ["<hobby or skill, 3-5 items>"],
             "persona": "<200-400 word detailed persona narrative>",
             "segment": "{segment_label}"

web/app.py

@@ -160,8 +160,8 @@ Return JSON:
     ]
 }}
 
-Make segments specific to THIS domain. For a product, use buyer personas. For a dating profile,
-use different types of potential partners. For a resume, use different hiring managers. Etc.
+Make segments specific to THIS domain. For a product, use buyer personas.
+For a resume, use different hiring managers. For a pitch, use different investor types. Etc.
 Be concrete and relevant — no generic segments."""
 
     try:

web/static/index.html

@@ -316,7 +316,7 @@
       <div class="step-num">1</div>
       <div class="step-title">What are you optimizing?</div>
     </div>
-    <p class="step-desc">Paste whatever you want evaluated — landing page, resume, pitch, dating profile, policy, anything.</p>
+    <p class="step-desc">Paste whatever you want evaluated — landing page, resume, pitch, policy, profile, anything.</p>
 
     <div class="template-chips">
       <span class="template-chip" onclick="loadTemplate('product')">Product</span>