Amazon Q and Amazon Bedrock, fully managed vs. custom - 2025-06-25
0 likes29 views
AIConf 2025 Overview of AWS services about machine learning and generative AI. https://www.aiconf.it/e/sessione/3727/Amazon-Q-and-Amazon-Bedrock-fully-managed-vs-custom
Amazon Q and Amazon Bedrock, fully managed vs. custom - 2025-06-25
- 1. Amazon Q and Bedrock, fully managed vs custom Alessandra Bilardi Data & Automation Specialist @ Corley Cloud >>AI CONF 2025
- 2. AI Conf 2025 Amazon Q and Bedrock, fully managed vs custom
- 3. AI Conf 2025 Oltre 500 progetti su AWS Corley Cloud è una realtà certificata con innumerevoli riconoscimenti e un portfolio di centinaia di progetti AWS sviluppati in diversi ambiti: cloud native, migrazione, machine learning & AI, serverless, IoT, sicurezza e cloudOps. Advanced Partner AWS
- 4. AI Conf 2025 Data & Automation Specialist @ Corley Cloud alessandra.bilardi@corley.it corley.it Alessandra Bilardi
- 5. AI Conf 2025 Alessandra Bilardi Data & Automation Specialist @ Corley Cloud alessandra.bilardi@corley.it corley.it
- 6. AI Conf 2025 Alessandra Bilardi Data & Automation Specialist @ Corley Cloud alessandra.bilardi@corley.it corley.it
- 7. AI Conf 2025 SUMMARY Machine learning steps and actors Generative AI with Amazon Q Generative AI with Amazon Bedrock Chat bot
- 8. AI Conf 2025 Machine learning steps and actors
- 9. AI Conf 2025 What are the steps of ML ? ➔ The data may arrive ready for learning, but often some processing is needed ➔ Model training could be delegated to an AI system, except for custom steps ➔ Evaluation is a prediction for which we know the expected values, for which we can calculate metrics ➔ The prediction works on new data processed with point 1 with the best model saved in point 3 Preparation Training & Tuning Testing & Evaluation Prediction (inference)
- 10. AI Conf 2025 ML system
- 11. AI Conf 2025 ML system
- 12. AI Conf 2025 ML system
- 13. AI Conf 2025 ML system
- 14. AI Conf 2025 ML
- 15. AI Conf 2025 ML
- 16. AI Conf 2025 Are there other steps or actors in ML ? ➔ Embeddings are objects that contain information about text, images, videos, audio or code ➔ The prompt is the text that contains the behavior that the model must have, the instructions to follow to respond to the request posed. ➔ Augmented Generation (AG) techniques allow us to exploit generalist models by providing them with instructions (the prompt), context (an extract of the embeddings) and a request to obtain a specific response. Question AG Answer Embeddings & Prompt LM
- 17. AI Conf 2025 Use case - Chat bot - preparation steps
- 18. AI Conf 2025 Use case - Chat bot - preparation steps
- 19. AI Conf 2025 Use case - Chat bot - preparation steps
- 20. AI Conf 2025 Use case - Chat bot - preparation steps
- 21. AI Conf 2025 Use case - Chat bot - preparation steps
- 22. AI Conf 2025 Use case - Chat bot - preparation steps
- 23. AI Conf 2025 Use case - Chat bot - preparation steps
- 24. AI Conf 2025 Use case - Chat bot - preparation steps
- 25. AI Conf 2025 Use case - Chat bot - preparation steps
- 26. AI Conf 2025 Use case - Chat bot - preparation steps
- 27. AI Conf 2025 Generative AI with Amazon Q
- 28. AI Conf 2025 Amazon Q
- 29. AI Conf 2025 Amazon Q
- 30. AI Conf 2025 Amazon Q
- 31. AI Conf 2025 Amazon Q Business
- 32. AI Conf 2025 Amazon Q Business 1. Embedding (as needed)
- 33. AI Conf 2025 Amazon Q Business 1. Embedding (as needed)
- 34. AI Conf 2025 Amazon Q Business 1. Embedding (as needed)
- 35. AI Conf 2025 Amazon Q Business 1. Embedding (as needed) 2. Request (question) 3. RAG 4. Response (answer)
- 36. AI Conf 2025 Amazon Q Business 1. Embedding (as needed) 2. Request (question) 3. RAG 4. Response (answer)
- 37. AI Conf 2025 Amazon Q Developer 1. Request
- 38. AI Conf 2025 Amazon Q Developer 1. Request 2. Embedding (as needed)
- 39. AI Conf 2025 Amazon Q Developer 1. Request 2. Embedding (as needed) 3. Context
- 40. AI Conf 2025 Generative AI with Amazon Bedrock
- 41. AI Conf 2025 Amazon Bedrock 1 2 Knowledge base (embedding) 3 Agent Models 4 Prompt
- 49. AI Conf 2025 Knowledge base
- 50. AI Conf 2025 Knowledge base
- 51. AI Conf 2025 Knowledge base
- 52. AI Conf 2025 Knowledge base
- 53. AI Conf 2025 Knowledge base
- 70. AI Conf 2025 Amazon Bedrock 1 2 Model evaluation 3 Playground Data automation 4 Prompt routers / caching
- 71. AI Conf 2025 Chat bot
- 72. AI Conf 2025 Use case - Chat bot - preparation steps
- 73. AI Conf 2025 Use case - Chat bot - preparation steps
- 74. AI Conf 2025 Use case - Chat bot - production version 1.0
- 75. AI Conf 2025 Use case - Chat bot - production version 1.0
- 76. AI Conf 2025 Use case - Chat bot - production version 1.0
- 77. AI Conf 2025 Use case - Chat bot - production version 1.0
- 78. AI Conf 2025 Use case - Chat bot - production version 1.0
- 79. AI Conf 2025 Use case - Chat bot - production version 1.1
- 80. AI Conf 2025 Use case - Chat bot - production version 2.0
- 81. AI Conf 2025 Use case - Chat bot - production version 2.0
- 82. AI Conf 2025 Use case - Chat bot - production version 2.0
- 83. AI Conf 2025 Use case - Chat bot - production version 2.0
- 84. AI Conf 2025 Use case - Chat bot - production version 2.0
- 85. AI Conf 2025 Which infrastructure for the ChatBot ?
- 86. AI Conf 2025 Solutions 1. like ChatGPT, max 30s
- 87. AI Conf 2025 Solutions 1. like ChatGPT, max 30s 2. extend 30s timeout
- 88. AI Conf 2025 Solutions 1. like ChatGPT, max 30s 2. extend 30s timeout Goals ● ↓ the response time ● ↓ the inference costs
- 89. AI Conf 2025 Inference ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2
- 90. AI Conf 2025 Inference ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2 ➔ Amazon SageMaker
- 91. AI Conf 2025 Inference ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2 ➔ Amazon SageMaker ➔ Amazon Bedrock / Q
- 92. AI Conf 2025 Inference Needs ➔ GPU ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2 ➔ Amazon SageMaker ➔ Amazon Bedrock / Q
- 93. AI Conf 2025 Inference Needs ➔ GPU ➔ Model loading ✖ ✖ ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2 ➔ Amazon SageMaker ➔ Amazon Bedrock / Q ✖
- 94. AI Conf 2025 Inference Needs ➔ GPU ➔ Model loading ✖ ✖ ➔ AWS Lambda ➔ Amazon Fargate ➔ Amazon EC2 ➔ Amazon SageMaker ➔ Amazon Bedrock / Q ✖
- 95. AI Conf 2025 Comparison of solutions
- 96. AI Conf 2025 AWS Services Comparison for a Chatbot Services Difficulty Embeddings Training $ Inference $ Amazon Q Business $0.264 / hour / 200MB $20 / user / mo Bedrock fine tuning $2 / 1000 queries $0.0079 / 1k tokens $30 / hour Bedrock on demand $0.00072 for input / 1k tokens and for output / 1k tokens Amazon SageMaker $2 / 1000 queries $0.921 / hour $0.921 / hour
- 97. AI Conf 2025 AWS Services Comparison for a Chatbot Services Difficulty Embeddings Training $ Inference $ Amazon Q Business $0.264 / hour / 200MB $20 / user / mo Bedrock fine tuning $2 / 1000 queries $0.0079 / 1k tokens $30 / hour Bedrock on demand $0.00072 for input / 1k tokens and for output / 1k tokens Amazon SageMaker $2 / 1000 queries $0.921 / hour $0.921 / hour
- 98. AI Conf 2025 AWS Services Comparison for a Chatbot Services Difficulty Embeddings Training $ Inference $ Amazon Q Business $0.264 / hour / 200MB $20 / user / mo Bedrock fine tuning $2 / 1000 queries $0.0079 / 1k tokens $30 / hour Bedrock on demand $0.00072 for input / 1k tokens and for output / 1k tokens Amazon SageMaker $2 / 1000 queries $0.921 / hour $0.921 / hour
- 99. AI Conf 2025 AWS Services Comparison for a Chatbot Services Difficulty Embeddings Training $ Inference $ Amazon Q Business $0.264 / hour / 200MB $20 / user / mo Bedrock fine tuning $2 / 1000 queries $0.0079 / 1k tokens $30 / hour Bedrock on demand $0.00072 for input / 1k tokens and for output / 1k tokens Amazon SageMaker $2 / 1000 queries $0.921 / hour $0.921 / hour
- 100. AI Conf 2025 Services Embeddings $ Training $ Inference $ Amazon Q Business 190 20 (user / mo) Bedrock fine tuning 2 1.5089 22320 Bedrock on demand 1.0714 (per 1k token) Amazon SageMaker 2 0.0154 3.68 (serverless) 685.22 (provisioned) AWS Services Comparison for a Chatbot Excluded from costs: ML storage, data processing and provisioned concurrency (serverless only) Example: 1 training of 191011 tokens + 1 request of 20s for every hour, every day for a month
- 101. AI Conf 2025 Services Embeddings $ Training $ Inference $ Amazon Q Business 190 20 (user / mo) Bedrock fine tuning 2 1.5089 22320 Bedrock on demand 1.0714 (per 1k token) Amazon SageMaker 2 0.0154 3.68 (serverless) 685.22 (provisioned) AWS Services Comparison for a Chatbot Excluded from costs: ML storage, data processing and provisioned concurrency (serverless only) Example: 1 training of 191011 tokens + 1 request of 20s for every hour, every day for a month
- 102. AI Conf 2025 Services Embeddings $ Training $ Inference $ Amazon Q Business 190 20 (user / mo) Bedrock fine tuning 2 1.5089 22320 Bedrock on demand 1.0714 (per 1k token) Amazon SageMaker 2 0.0154 3.68 (serverless) 685.22 (provisioned) AWS Services Comparison for a Chatbot Excluded from costs: ML storage, data processing and provisioned concurrency (serverless only) Example: 1 training of 191011 tokens + 1 request of 20s for every hour, every day for a month
- 103. AI Conf 2025 Services Embeddings $ Training $ Inference $ Amazon Q Business 190 20 (user / mo) Bedrock fine tuning 2 1.5089 22320 Bedrock on demand 1.0714 (per 1k token) Amazon SageMaker 2 0.0154 3.68 (serverless) 685.22 (provisioned) AWS Services Comparison for a Chatbot Excluded from costs: ML storage, data processing and provisioned concurrency (serverless only) Example: 1 training of 191011 tokens + 1 request of 20s for every hour, every day for a month
- 104. AI Conf 2025 Services Embeddings $ Training $ Inference $ Amazon Q Business 190 Bedrock fine tuning 2 1.5089 22320 Bedrock on demand 1.0714 (per 1k token) Amazon SageMaker 2 0.0154 3.68 (serverless) 685.22 (provisioned) AWS Services Comparison for a Chatbot Excluded from costs: ML storage, data processing and provisioned concurrency (serverless only) Example: 1 training of 191011 tokens + 1 request of 20s for every hour, every day for a month
- 105. AI Conf 2025 Thanks for listening!
- 106. Thank you! >>AI CONF 2025 👉 slides & videos: https://www.improove.tech/videos