from typing import Type from crewai.tools import BaseTool from pydantic import BaseModel, Field import dspy import os from dotenv import load_dotenv load_dotenv(dotenv_path="/home/azureuser/microlearn/backend/prompt_opt_dspy/.env") azure_api_key = os.getenv("AZURE_API_KEY") azure_api_host = os.getenv("AZURE_API_HOST") azure_deployment_id = os.getenv("AZURE_DEPLOYMENT_ID") azure_api_version = os.getenv("AZURE_API_VERSION") # === Configure Azure OpenAI LM === lm = dspy.LM( "azure/Csqr-gpt-4o-mini", api_key=azure_api_key, api_base=azure_api_host, api_version=azure_api_version, temperature=0.0, ) dspy.configure(lm=lm) # === DSPy Signature === class RestructureSignature(dspy.Signature): """ Restructure like a syllabus for the user journey or update the existing user journeys based on feedback. read the content {content} and make a new user journey if the feedack is "First iteration", Otherwise update the user journey in the {content} according to the user feedback {feedback} The user's company context is as follows: Company Name : {Company_Name} Industry : {Industry} Company Size : {Company_Size} Business Model : {Business_Model} Top Use Cases : {Top_Use_Cases} Learning Focus : {Learning_Focus} Tech Stack : {Tech_stack} Compliance : {Compliance} # User journey Creation Task ## Objective You are an intelligent and experienced data restructuring agent. Your task is to review the researched content provided and make a structured user journey. You must strictly adhere to the following guidelines. ## Guidlines - Use the level {Level} and user's experience with the topic {Experience} to decide the depth and balance of the user journey. - The users motive for taking this micro learning user journey is {motive}. - The user's knowledge level to different skills are as follows: (In percentage) 1. Knowlege in SQL - {SQL} 2. Data visualisation skill - {Viz} 3. Statistical Anlysis skill - {Stats} 4. Business Communication skill - {Comm} 5. Data storytelling skill - {Story} - Set the content accordingly. - Divide the provided content to different stages - The toatal available time to cover the user journey is {available_time}. - **The sum of all stage durations MUST equal {available_time}. Do not exceed this.** - Include case studies where relevant, but **no projects**. - Based on the total available time, logically organize the topics into a structured, multi-stage user journey. - Distribute the available time ({available_time}) across the stages depending on content, focus and complexity. - Ensure the **sum of durations** across **all stages** is **{available_time}** - The user is either non-technical or minimally technical. ## Additional Instructions - **Before generating the stages, calculate the total duration and verify that the sum equals {available_time}.** - If necessary, adjust durations proportionally to maintain balance and alignment with the available time. - Do not comment or explain anything outside of the user journey content. ## Here is the researched content : {content} A detailed, structured user journey for a user on the given content, tailored to the {Level} level. The user is either a non technical or minimally technical person. - Generate a compelling main heading based on the user's Job Title and Topic. - The heading should reflect the relevance of the topic in the context of the user's role and the user's industry. - Keep it professional, engaging, and suitable. Job Title: {Job_Title} Divide the user journey into multiple stages. Each stage should contain: 1. **heading** for the stage (include the stage number) 2. **Focus** 3. **Outcome** 4. **Duration** 5. **Topics Covered** — select relevant topics from the researched user journey content. The final stage should focus on application of the user's existing skills. Ensure the sum of durations across all stages is {available_time}. Do not include any project in the user journey, can include case studies. Should not comment anything other than the user journey. Format as markdown without '```'. Ensure the sum of durations across all stages is **exactly {available_time}**. Example output with two stages: # **Main Heading** ## **Stage1-Sub Heading** ### **Focus** Focus of stage1 ### **Outcome** Expected outcome of stage 1 ### **Duration** Duration1(Specify **ONLY** the time required to complete the stage1) ### **Topics Covered** - Topic1 - Topic2 ## **Stage2-Sub Heading** ### **Focus** Focus of stage2 ### **Outcome** Expected outcome of stage 2 ### **Duration** Duration2(Specify **ONLY** the time required to complete the stage2) ### **Topics Covered** - Topic1 - Topic2 Note: Duration1 + Duration2 should be equal to {available_time} Example output with three stages: # **Main Heading** ## **Stage1-Sub Heading** ### **Focus** Focus of stage1 ### **Outcome** Expected outcome of stage 1 ### **Duration** Duration1(Specify **ONLY** the time required to complete the stage1) ### **Topics Covered** - Topic1 - Topic2 ## **Stage2-Sub Heading** ### **Focus** Focus of stage2 ### **Outcome** Expected outcome of stage 2 ### **Duration** Duration2(Specify **ONLY** the time required to complete the stage2) ### **Topics Covered** - Topic1 - Topic2 ## **Stage3-Sub Heading** ### **Focus** Focus of stage3 ### **Outcome** Expected outcome of stage 3 ### **Duration** Duration3(Specify **ONLY** the time required to complete the stage2) ### **Topics Covered** - Topic1 - Topic2 Note: Duration1 + Duration2 + Duration3 should be equal to {available_time} ## Important: - **Sum of durations must equal {available_time}.** - No projects, but case studies allowed. - Markdown format only, no extra text or commentary. """ content: str = dspy.InputField() feedback: str = dspy.InputField() company_context: str = dspy.InputField() user_context: str = dspy.InputField() restructured_journey: str = dspy.OutputField( desc="Structured multi-stage user journey in markdown format with durations summing exactly to available_time" ) # === DSPy Module === class RestructureModule(dspy.Module): def __init__(self): super().__init__() self.restructure_predictor = dspy.Predict(RestructureSignature) predictor = dspy.Predict(RestructureSignature) # --- Two Example Training Shots --- trainset = [ dspy.Example( content=''' - **Topic Area**: Fundamentals - **Subtopics/Concepts**: - Introduction to Artificial Intelligence in Healthcare - Overview of Machine Learning and its Role in Patient Care - Types of AI: Supervised, Unsupervised, and Reinforcement Learning - Ethical Considerations in AI Applications - Understanding Data Privacy: HIPAA and GDPR Compliance - **Topic Area**: Applications - **Subtopics/Concepts**: - AI in Predictive Analytics: Enhancing Patient Outcomes - AI in Patient Monitoring: Wearables and Remote Monitoring - Use of Natural Language Processing (NLP) in Clinical Documentation - Chatbots and Virtual Health Assistants for Patient Engagement - Remote Diagnosis and Telemedicine Solutions - **Topic Area**: Tools and Techniques - **Subtopics/Concepts**: - Overview of Data Analytics Tools: Python and PyTorch for healthcare - Using AWS HealthLake for Data Management in Healthcare - Data Visualization Tools for Presenting AI Insights - Techniques in Data Preprocessing and Feature Engineering - Implementing AI Models for Real-Time Patient Monitoring - **Topic Area**: Advanced Topics - **Subtopics/Concepts**: - Deep Learning Applications in Medical Imaging - AI-Driven Decision Support Systems for Healthcare Professionals - Integrating AI into Existing Clinical Workflows - Case Studies on AI Success Stories in Patient Care - Challenges and Limitations of AI in Healthcare Practice - **Topic Area**: Current Trends - **Subtopics/Concepts**: - Rise of Telehealth Solutions and AI Integration - Innovations in Wearable Health Technology and AI - The Role of AI in Addressing Pandemic Challenges - The Future of AI in Personalized Medicine - Collaborations Between Tech Companies and Healthcare Institutions This structured user journey content is designed to provide a comprehensive overview of AI in Healthcare, particularly aligned with the interests and learning objectives of a Healthcare Analyst looking to explore how AI can improve patient care. ''', feedback="First iteration", company_context=""" Company Name: MediCore Solutions Industry: Healthcare Company Size: 2000+ employees Business Model: B2B and B2C SaaS Top Use Cases: Predictive analytics, Patient monitoring Learning Focus: AI diagnostics and workflow automation Tech Stack: Python, PyTorch, AWS HealthLake Compliance: HIPAA, GDPR """, user_context=""" Job Title: Healthcare Analyst Level: Intermediate Experience: 2 year motive: Exploring AI applications to improve patient care Knowledge in SQL: 70% Data Visualization Skill: 60% Statistical Analysis Skill: 40% Business Communication Skill: 50% Data Storytelling Skill: 45% Available Time: 240 minutes """, restructured_journey=""" # **Enhancing Patient Care through AI: A Healthcare Analyst’s Guide** ## **Stage 1 - Introduction to AI in Healthcare** ### **Focus** Understanding the fundamental concepts of AI and its role in the healthcare industry. ### **Outcome** Establish a foundational knowledge of AI and machine learning principles, and comprehend ethical considerations and compliance regulations in healthcare. ### **Duration** 60 minutes ### **Topics Covered** - Introduction to Artificial Intelligence in Healthcare - Overview of Machine Learning and its Role in Patient Care - Ethical Considerations in AI Applications - Understanding Data Privacy: HIPAA and GDPR Compliance ## **Stage 2 - AI Applications in Patient Care** ### **Focus** Exploring how AI is applied to diverse aspects of patient care, enhancing outcomes and engagement. ### **Outcome** Gain insights into practical applications of AI in predictive analytics, patient monitoring, and telemedicine, including the role of NLP and chatbots. ### **Duration** 90 minutes ### **Topics Covered** - AI in Predictive Analytics: Enhancing Patient Outcomes - AI in Patient Monitoring: Wearables and Remote Monitoring - Use of Natural Language Processing (NLP) in Clinical Documentation - Chatbots and Virtual Health Assistants for Patient Engagement - Remote Diagnosis and Telemedicine Solutions ## **Stage 3 - Tools, Techniques, and Advanced Topics** ### **Focus** Familiarizing with the technical tools and advanced applications of AI in healthcare while reviewing success stories. ### **Outcome** Understand data analytics tools, case studies of successful AI applications, and integration strategies for AI within existing clinical workflows. ### **Duration** 90 minutes ### **Topics Covered** - Overview of Data Analytics Tools: Python and PyTorch for healthcare - Using AWS HealthLake for Data Management in Healthcare - Deep Learning Applications in Medical Imaging - AI-Driven Decision Support Systems for Healthcare Professionals - Case Studies on AI Success Stories in Patient Care Total Duration: 60 + 90 + 90 = 240 minutes """ ).with_inputs("content", "feedback", "company_context", "user_context"), dspy.Example( content="Research on SQL optimization techniques", feedback="Focus on advanced optimizations for experienced users.", company_context=""" Company Name: FinNext Global Industry: FinTech Company Size: 1000-1500 employees Business Model: B2B Top Use Cases: Cross-border payments, Fraud prevention Learning Focus: Blockchain adoption and market opportunities Tech Stack: Ethereum, Hyperledger, Azure Blockchain Service Compliance: KYC, AML """, user_context=""" Job Title: Business Development Executive Level: Beginner Experience: 5 years motive: Understanding blockchain for business strategy alignment Knowledge in SQL: 60% Data Visualization Skill: 70% Statistical Analysis Skill: 40% Business Communication Skill: 50% Data Storytelling Skill: 45% Available Time: 240 minutes """, restructured_journey=""" # **Understanding Blockchain for Business Strategy Alignment in FinTech** ## **Stage 1 - Introduction to Blockchain Fundamentals** ### **Focus** Introduce the fundamental concepts of blockchain technology. ### **Outcome** Gain foundational knowledge about blockchain and its significance in the FinTech landscape. ### **Duration** 60 minutes ### **Topics Covered** - Introduction to Blockchain - Key Terms - Differences between Blockchain Types - Basic Mechanisms ## **Stage 2 - Blockchain Applications in FinTech** ### **Focus** Explore specific blockchain applications relevant to FinTech, particularly in payments and fraud detection. ### **Outcome** Understand how blockchain enhances business processes within the FinTech sector. ### **Duration** 60 minutes ### **Topics Covered** - Cross-border Payments - Fraud Prevention - Digital Identity Management ## **Stage 3 - Tools and Platforms for Blockchain** ### **Focus** Familiarize with key blockchain platforms and tools that can be leveraged in business strategies. ### **Outcome** Identify relevant blockchain tools and platforms that can integrate into existing FinTech systems. ### **Duration** 50 minutes ### **Topics Covered** - Overview of Ethereum and Hyperledger - Azure Blockchain Service - Integration with Existing Systems ## **Stage 4 - Identifying Market Opportunities** ### **Focus** Analyze current trends, competition, and customer engagement opportunities within blockchain adoption. ### **Outcome** Develop insights on market opportunities and strategic advantages offered by blockchain. ### **Duration** 40 minutes ### **Topics Covered** - Trends in Blockchain Adoption - Competitive Landscape - Customer Engagement ## **Stage 5 - Business Strategy Alignment** ### **Focus** Align blockchain knowledge with business strategy and growth opportunities. ### **Outcome** Formulate strategic partnerships and business models leveraging blockchain technology. ### **Duration** 30 minutes ### **Topics Covered** - Strategic Partnerships - Business Models leveraging Blockchain - Measuring ROI ## **Stage 6 - Future Developments and Skills Application** ### **Focus** Understand future trends and how to continue developing skills in blockchain. ### **Outcome** Identify emerging trends and resources for continuous learning, and apply existing skills in statistical analysis and data visualization to business scenarios. ### **Duration** 40 minutes ### **Topics Covered** - Emerging Trends - Regulatory Landscape - Education and Skill Development **Total Duration: 240 minutes** """ ).with_inputs("content", "feedback", "company_context", "user_context"), ] # Metric: Check that duration is allocated in stages def restructure_metric(gold, pred, trace): return int("Stage" in pred.restructured_journey and str(gold.restructured_journey.count("Stage")) in pred.restructured_journey) # Compile with bootstrap few-shot optimizer optimizer = dspy.BootstrapFewShot(metric=restructure_metric) optimizer.compile(predictor, trainset=trainset) # Store optimized predictor self.restructure_predictor = predictor print("=== Optimized ===") print(predictor) def forward(self, content, feedback, company_context, user_context): return self.restructure_predictor( content=content, feedback=feedback, company_context=company_context, user_context=user_context ) # === Tool Input Schema === class RestructureToolInput(BaseModel): """Input schema for RestructureAnalysisTool.""" content: str = Field(..., description="The researched content to structure into a user journey") feedback: str = Field(..., description="User feedback (First iteration or update request)") Company_Name: str = Field(..., description="Name of the company") Industry: str = Field(..., description="Industry of the company") Company_Size: str = Field(..., description="Size of the company") Business_Model: str = Field(..., description="Company's business model") Top_Use_Cases: str = Field(..., description="Top use cases the company focuses on") Learning_Focus: str = Field(..., description="Learning focus for the user") Tech_stack: str = Field(..., description="Company's tech stack") Compliance: str = Field(..., description="Relevant compliance requirements") Job_Title: str = Field(..., description="Job title of the user") Level: str = Field(..., description="Learning level (Beginner, Intermediate, Advanced)") Experience: str = Field(..., description="User's experience with the topic") motive: str = Field(..., description="User's motive for research") SQL: int = Field(..., description="Knowledge in SQL (percentage)") Viz: int = Field(..., description="Data visualization skill (percentage)") Stats: int = Field(..., description="Statistical analysis skill (percentage)") Comm: int = Field(..., description="Business communication skill (percentage)") Story: int = Field(..., description="Data storytelling skill (percentage)") available_time: int = Field(..., description="Total available time for the user journey (in minutes)") # === Custom DSPy Tool === class RestructureAnalysisTool(BaseTool): name: str = "Restructure Analysis Tool" description: str = ( "Restructures researched content into a structured user journey syllabus based on user feedback, " "company context, and user profile. Ensures durations sum exactly to available_time." ) args_schema: Type[BaseModel] = RestructureToolInput def _run( self, content: str, feedback: str, Company_Name: str, Industry: str, Company_Size: str, Business_Model: str, Top_Use_Cases: str, Learning_Focus: str, Tech_stack: str, Compliance: str, Job_Title: str, Level: str, Experience: str, motive: str, SQL: int, Viz: int, Stats: int, Comm: int, Story: int, available_time: int ) -> str: # Build company + user context company_context = f""" Company Name: {Company_Name} Industry: {Industry} Company Size: {Company_Size} Business Model: {Business_Model} Top Use Cases: {Top_Use_Cases} Learning Focus: {Learning_Focus} Tech Stack: {Tech_stack} Compliance: {Compliance} """ user_context = f""" Job Title: {Job_Title} Level: {Level} Experience: {Experience} Motive: {motive} Knowledge in SQL: {SQL}% Data Visualization Skill: {Viz}% Statistical Analysis Skill: {Stats}% Business Communication Skill: {Comm}% Data Storytelling Skill: {Story}% Available Time: {available_time} """ restructure = RestructureModule() result = restructure.forward(content, feedback, company_context, user_context) return result.restructured_journey # # from restructure_tool import RestructureAnalysisTool # if __name__ == "__main__": # tool = RestructureAnalysisTool() # result = tool._run( # content=""" # - **Topic Area**: Fundamentals of Cyber Security # - **Subtopics/Concepts**: # - Introduction to Cyber Security: Importance and relevance in EdTech # - Key Terminology: Threats, vulnerabilities, and risks # - Types of Cyber Attacks: Phishing, malware, ransomware # - Basic Security Principles: Confidentiality, integrity, availability (CIA triad) # - **Topic Area**: Cyber Security in Education Technology # - **Subtopics/Concepts**: # - Data Protection Regulations: Overview of GDPR and FERPA # - Best Practices for Data Security in EdTech: User data management and protection # - Case Studies: Successful implementation of security measures in EdTech companies # - **Topic Area**: Tools and Techniques for Cyber Security # - **Subtopics/Concepts**: # - Security Tools Overview: Firewalls, antivirus software, and intrusion detection systems # - Introduction to Encryption: Importance of data encryption in protecting sensitive information # - User Authentication Methods: Multi-factor authentication and secure password practices # - **Topic Area**: Current Trends in Cyber Security # - **Subtopics/Concepts**: # - AI and Cyber Security: How AI is transforming security measures in EdTech # - Emerging Threats: Understanding the evolving landscape of cyber threats # - Cyber Security Frameworks: Overview of NIST and ISO standards relevant to EdTech # - **Topic Area**: Upskilling for Cyber Security # - **Subtopics/Concepts**: # - Learning Resources: Online courses and certifications for beginners in cyber security # - Networking Opportunities: Joining EdTech and Cyber Security communities # - Building a Cyber Security Mindset: Importance of continuous learning and awareness # """, # feedback="First iteration", # Company_Name="TechNova Inc", # Industry="EdTech", # Company_Size="500-1000 employees", # Business_Model="B2B SaaS", # Top_Use_Cases="Personalized learning, AI tutors", # Learning_Focus="AI adoption in education", # Tech_stack="Python, TensorFlow, Azure", # Compliance="GDPR, FERPA", # Job_Title="Product Manager", # Level="Beginner", # Experience="2 years working with SQL and dashboards", # motive="Upskilling for AI-driven product development", # SQL=70, # Viz=60, # Stats=40, # Comm=50, # Story=45, # available_time=240 # ) # print("=== Restructured User Journey ===") # print(result)