Bridging Healthcare Gap: GenAI Therapist with human empathy (Part 6)

Danny Chan - Sep 14 - - Dev Community

Old School Project Management:

Image description

Step by step to create a project
Take a long development time and resources
Without real-time feedback from the target user
Solution may not solve the real problem
People tend to give up because there is no upfront value
Build without trust and confidence
The market is always changing, so solutions become outdated.

Step 1: Document
Step 2: Development
Step 3: API integration
Step 4: Testing
Step 5: Launch product
Step 6: Gain user feedback
Step 7: Optimize feature



What Modern Hackers Should Do:

Image description

Work in parallel on every step to create value and experiences

  • Short development time

Spend fewer resources to create a workable minimal viable product

  • Gain real-time feedback from the target user Keep tracking if the solution is solving the real-world problem
  • Provide upfront value to all stakeholders

Build with trust and confidence

  • Market is always changing, so always upgrade products to fulfill user needs.


Good Project Ideas:

Good project ideas come from your daily life.

(1) Be humble

(2) Understand other people's situations, obstacles, and feelings
What are the pain points within society?

(3) Define the problem and root cause
Are we solving the right problem?

(4) Iterate the solution and create a feedback loop
Can a solution solve the problem?

(5) Analyze data from the feedback loop to realize the customer's real needs
Are we using the right tools to solve the right problem?

(6) Upgrade the system, optimize to fulfill user needs
Are we providing good products/services to solve problems?



Good Design:

The good design process can be flexibly adjusted according to the problem
Rather than letting the problem adapt to the process itself



VUCA:

Technological innovation rapidly changes industry and lifestyle

Volatility:

  • The rapid and unpredictable nature of change

Uncertainty:

  • The unpredictability of events and issues

Complexity:

  • The unclear cause-and-effect relationship

Ambiguity:

  • The potential for misunderstandings from mixed messages

The U.S. Army War College introduced the concept of VUCA in 1987



We are all builders

We have the capacity, desire to learn, grow skills
To build something amazing
Iterate fast, and optimize
Provide customers with faster service
Explore revenue channels



Key Takeaway:

Being an innovative GenAI Builder
Solve real-world problems through GenAI
Empowered by AWS Bedrock to build ideas
Embrace change to create a startup



More materials:

https://serverlessland.com/

Pattern

  • Bedrock + Lambda
  • Connect to Amazon Bedrock using AWS Lambda

Workflow

  • Bedrock + Lambda + Gateway
  • Amazon Bedrock Serverless Chatbot Workshop

Repository

  • S3 + Bedrock + Lambda
  • Generate blog from images using Bedrock Claude v3 sonnet


More materials:

Prompt templates and examples for Amazon Bedrock text models
https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-templates-and-examples.html

Text classification
Question-answer, without context
Question-answer, with context
Summarization
Text generation
Code generation
Mathematics
Reasoning/logical thinking
Entity extraction
Chain-of-thought reasoning



Reference:

Separating Data and Instructions
https://catalog.us-east-1.prod.workshops.aws/workshops/0644c9e9-5b82-45f2-8835-3b5aa30b1848/en-US/lessons/lab-4-separating-data-and-instructions

Few-Shot Learning
https://catalog.us-east-1.prod.workshops.aws/workshops/a4bdb007-5600-4368-81c5-ff5b4154f518/en-US/050-prompt-engineering

Lab 8 - Creating Agent with Guardrails for Amazon Bedrock integration
https://catalog.workshops.aws/agents-for-amazon-bedrock/en-US/80-create-agent-with-guardrails

Guardrails for Amazon Bedrock Workshop
https://catalog.us-east-1.prod.workshops.aws/workshops/0720c7c4-fb23-4e43-aa9f-036fc07f46b2

Creating Responsible AI with Amazon Bedrock Guardrails
https://catalog.workshops.aws/bedrockguard/en-US/amazon-bedrock-guardrails

Build and scale generative AI applications with Amazon Bedrock
https://catalog.us-east-1.prod.workshops.aws/workshops/e820beb4-e87e-4a85-bc5b-01548ceba1f8/en-US/lab-1-extract-insights-using-prompt-engineering-techniques/prompt-engineering-basics#inference-parameters

Amazon Bedrock Workshop
https://catalog.us-east-1.prod.workshops.aws/workshops/a4bdb007-5600-4368-81c5-ff5b4154f518/en-US

Prompt Engineering with Anthropic's Claude 3
https://catalog.us-east-1.prod.workshops.aws/workshops/0644c9e9-5b82-45f2-8835-3b5aa30b1848/en-US/lessons/lab-9-complex-prompts-from-scratch#lesson-objective:

Caring for Your Mental Health
https://www.nimh.nih.gov/health/topics/caring-for-your-mental-health

Design solutions to problems with the Double Diamond process
https://www.bitesizelearning.co.uk/resources/double-diamond-design-process-explained

Lean entrepreneur
https://leanentrepreneur.co/illustrations

Icon Pack: Avatar Compilation
https://www.flaticon.com/packs/avatar-compilation

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .