counter

app.py

@@ -126,6 +126,10 @@ If your product discovery experience isn’t working as hard as your marketing b
     ================================================
     - Retrieval: Public Product Data using Tavily Search
     - Recommend: Competition Product 
+    
+    ### benefits
+    - remove friction in research, saving labour time
+    - improve insight quality by identify competitor
     """)
     in_verbatim = gr.Textbox(label="Verbatim")
     out_product = gr.Textbox(label="Product")
@@ -237,6 +241,7 @@ Uses customer data and behavior to craft messages that resonate with specific se
     gr.Markdown("""
 Objective: Transform Personal Pain Points into Actionable Insights with a Dynamic Knowledge Graph Framework
 =====================================
+- Identify what channel customer prefer
     """)
     in_verbatim = gr.Textbox(label="Question")
     out_product = gr.JSON(label="Knowledge Graph")
@@ -371,13 +376,18 @@ Allows downstream tasks (like sentiment analysis or topic modeling) to focus on
     """)
 
 
-  with gr.Tab("Classify"):
+  with gr.Tab("Segmentation"):
     gr.Markdown("""
     Objective: Streamline Customer Insights: Auto-Classify Feedback for Product Optimization
     ================================================
     - multi class classification, could have multiple label for 1 feedback
     - fix classification in this use case: online banking, card, auto finance, mortgage, insurance
     - LLM Judge to evaluate relevancy
+    
+    Business use case: customer segmentation for ab testing
+    ------------------------------------------------
+      - Acquisition: Behavior cluster, we can predict not only who is likely to click—but who is likely to retain
+      - Activation: segmenting users based on behavioral signals—like browsing activity, time since last engagement, or declining open/click rates.
     """)
     in_verbatim = gr.Textbox(label="Customer Feedback separate by ;")
     out_product = gr.Textbox(label="Classification & Evaluation")
@@ -506,6 +516,7 @@ For example, Comcast reduced repeat service calls by 17% after deploying entity
     gr.Markdown("""
     Objective: Leveraging Human Feedback to Deliver Personalized Content that Proactively Solves Customer Pain Points
     ================================================
+    - replace human with reward/penalty function, you will get RLHF by ranking the solutions
     """)
 
     in_verbatim = gr.Textbox(label="Persona")

knowledge.py

@@ -88,7 +88,7 @@ def generate_graph(q, input=KnowledgeGraph()) -> KnowledgeGraph:
     messages=[
       {
         "role": "user",
-        "content": dedent(f"""As a world class iterative knowledge graph builder and a Marketing Data Scientist for delivery personalized solution in Personal and Commercial Banking. Help me understand this person pain points and needs by describing the interaction as a detailed knowledge graph: 
+        "content": dedent(f"""As a world class iterative knowledge graph builder and a Marketing Data Scientist for delivery personalized solution in Personal and Commercial Banking. Help me understand this person pain points, preference of contacts and needs by describing the interaction as a detailed knowledge graph: 
             ### Interaction: {q} 
             ### Merge from existing KnowledgeGraph, Here is the current state of the graph:
             {input.model_dump_json()}

rag.py

@@ -4,7 +4,8 @@ import pandas as pd
 
 #lm = dspy.LM('ollama_chat/deepseek-r1', api_base='http://localhost:11434', api_key='')
 #lm = dspy.LM('huggingface/Qwen/Qwen2.5-Coder-32B-Instruct')
-lm = dspy.LM('huggingface/meta-llama/Llama-3.2-1B')
+#lm = dspy.LM('huggingface/meta-llama/Llama-3.2-1B')
+lm = dspy.LM('groq/gemma-7b-it')
 dspy.configure(lm=lm)
 
 df = pd.read_csv("product2.csv")

tool.py

@@ -6,7 +6,8 @@ from tavily import TavilyClient
 
 #lm = dspy.LM('ollama_chat/deepseek-r1', api_base='http://localhost:11434', api_key='')
 #lm = dspy.LM('huggingface/Qwen/Qwen2.5-Coder-32B-Instruct')
-lm = dspy.LM('huggingface/meta-llama/Llama-3.2-1B')
+#lm = dspy.LM('huggingface/meta-llama/Llama-3.2-1B')
+lm = dspy.LM('groq/gemma-7b-it')
 dspy.configure(lm=lm)
 
 search_client = TavilyClient(api_key=os.environ["T_TOKEN"])