app.py

app.py

@@ -0,0 +1,94 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+import faiss
+from openai import OpenAI
+
+client = OpenAI()
+
+def row_to_text(row):
+    return (
+        f"Clinic: {row['Account_Name']}\n"
+        f"Region: {row['Region']}\n"
+        f"Account Manager: {row['Account_Manager']}\n"
+        f"Revenue: €{row['Revenue']}\n"
+        f"Last Purchase: {row['Last_Purchase_Date']}\n"
+        f"Churn Risk: {row['Churn_Risk']}\n"
+        f"Satisfaction Score: {row['Satisfaction_Score']}/10\n"
+        f"Product Interest: {row['Product_Interest']}\n"
+        f"Notes: {row['Account_Notes']}"
+    )
+
+def get_embedding(text):
+    response = client.embeddings.create(
+        input=[text],
+        model="text-embedding-3-small"
+    )
+    return response.data[0].embedding
+
+def contextual_pitch_assistant(csv_file, query):
+    df = pd.read_csv(csv_file.name)
+    text_chunks = df.apply(row_to_text, axis=1).tolist()
+    embeddings = [get_embedding(t) for t in text_chunks]
+
+    dim = len(embeddings[0])
+    index = faiss.IndexFlatL2(dim)
+    index.add(np.array(embeddings).astype('float32'))
+
+    q_emb = np.array([get_embedding(query)]).astype('float32')
+    D, I = index.search(q_emb, 3)
+    retrieved = [text_chunks[i] for i in I[0]]
+
+    prompt = f"""
+    You are an expert in B2B sales and digital pitching for dental clinics.
+    Based on the following CRM insights and query, generate:
+    1) A short HTML email pitch (subject + body) ready to send.
+    2) A DALL·E 3 prompt for a matching header image.
+
+    Query:
+    {query}
+
+    CRM data:
+    {'---'.join(retrieved)}
+    """
+
+    response = client.chat.completions.create(
+        model="gpt-4o-mini",
+        messages=[{"role": "user", "content": prompt}],
+        temperature=0.7
+    )
+
+    output_text = response.choices[0].message.content
+
+    import re
+    match = re.search(r"(Image prompt|DALL·E prompt)[:\-]\s*(.*)", output_text)
+    image_prompt = match.group(2).strip() if match else "modern dental clinic interior with dentist and patient"
+
+    image = client.images.generate(
+        model="gpt-image-1",
+        prompt=image_prompt,
+        size="1024x1024"
+    )
+    image_url = image.data[0].url
+
+    html = f"""
+    <div style='font-family:Arial,sans-serif;max-width:600px;margin:auto;padding:16px;background:#ffffff;
+    border-radius:12px;box-shadow:0 2px 8px rgba(0,0,0,0.1);'>
+      <img src="{image_url}" style="width:100%;border-radius:8px;margin-bottom:16px;">
+      {output_text}
+    </div>
+    """
+    return html
+
+app = gr.Interface(
+    fn=contextual_pitch_assistant,
+    inputs=[
+        gr.File(label="Upload CRM CSV (5–100 rows)", file_types=[".csv"]),
+        gr.Textbox(label="Sales Query", placeholder="e.g. Which clinic is best for our imaging subscription?")
+    ],
+    outputs=gr.HTML(label="Email Pitch Preview"),
+    title="Contextual Pitch Assistant for Dental Sales",
+    description="Upload CRM data and get a personalized contextual pitch with generated imagery."
+)
+
+app.launch()