Building Intelligent Troubleshooting Agents
Keep adding new skills with 10,000+ programs for $239 (usually $399). Save now.
Building Intelligent Troubleshooting Agents
This course is part of Microsoft AI & ML Engineering Professional Certificate
Instructor: Microsoft
7,265 already enrolled
Included with
Learn more
Ask Coursera
19 reviews
Recommended experience
19 reviews
Recommended experience
Skills you'll gain
- Decision Support Systems
- Agentic systems
- Decision Intelligence
- Fine-tuning
- Model Evaluation
- Large Language Modeling
- Model Training
- Artificial Intelligence
- Performance Testing
- Model Optimization
- Generative AI Agents
- Natural Language Processing
- User Interface (UI)
- Machine Learning Algorithms
- Applied Machine Learning
- LLM Application
- Test Case
Tools you'll learn
Details to know
See how employees at top companies are mastering in-demand skills
Build your Software Development expertise
- Learn new concepts from industry experts
- Gain a foundational understanding of a subject or tool
- Develop job-relevant skills with hands-on projects
- Earn a shareable career certificate from Microsoft
There are 5 modules in this course
This course focuses on the design and implementation of intelligent troubleshooting agents. You will learn to create AI-powered agents that can diagnose and resolve issues autonomously. The course covers natural language processing, decision-making algorithms, and best practices in AI agent development.
By the end of this course, you will be able to: 1. Define, describe, and design the architecture of an intelligent troubleshooting agent. 2. Implement natural language processing techniques for user interaction. 3. Develop decision-making algorithms for problem diagnosis and resolution. 4. Optimize and evaluate the performance of AI-based troubleshooting agents. To be successful in this course, you should have intermediate programming knowledge of Python, plus experience with AI & ML infrastructure and core algorithms and techniques, including approaches using pretrained large-language models (LLMs). Familiarity with statistics is also recommended.
In this module, you'll delve into the critical processes and methodologies involved in fine-tuning LLMs to enhance their performance for specific tasks. By the end of this module, you will have a comprehensive understanding of fine-tuning techniques and be equipped to apply these methods to enhance LLMs for specific, practical applications.
What's included
11 videos29 readings13 assignments
11 videosβ’Total 66 minutes
- Introduction to the AI/ML engineering advanced professional certificate programβ’4 minutes
- Introduction to LLM fine-tuning for task-specific adaptationβ’4 minutes
- The importance of fine-tuning an LLMβ’4 minutes
- Walkthrough: Creating your code repository Part 1 (Optional)β’5 minutes
- Walkthrough: Creating your code repository Part 2 (Optional)β’8 minutes
- Use case demonstration: Selecting and preparing data for fine-tuningβ’7 minutes
- Walkthrough: Preparing a dataset for fine-tuning (Optional)β’5 minutes
- Walkthrough: Comparing fine-tuning techniques (Optional)β’13 minutes
- The relevance of evaluation metricsβ’5 minutes
- Summary: Fine-tuning LLMsβ’4 minutes
- Walkthrough: Fine-tuning an LLM (Optional)β’7 minutes
29 readingsβ’Total 724 minutes
- Welcome to the Coursera Communityβ’2 minutes
- Microsoft updatesβ’2 minutes
- Practice activity: Setting up your environment in Microsoft Azureβ’30 minutes
- Walkthrough: Setting up your environment in Microsoft Azure (Optional)β’0 minutes
- Practice activity: Creating your code repositoryβ’60 minutes
- Course syllabusβ’10 minutes
- Overview of LLM fine-tuningβ’10 minutes
- LLM Fine-Tuning: Principles and Stepsβ’10 minutes
- Practice activity: LLM fine-tuningβ’30 minutes
- Walkthrough: LLM fine-tuning (Optional)β’0 minutes
- Detailed explanation of principles and steps of LLM fine-tuningβ’10 minutes
- Review: Principles and steps of LLM fine-tuningβ’15 minutes
- Selecting and preparing data for fine-tuningβ’10 minutes
- Practice activity: Model and dataset selectionβ’30 minutes
- Walkthrough: Model and dataset selection (Optional)β’0 minutes
- Practice activity: Preparing a dataset for fine-tuningβ’60 minutes
- Fine-tuning techniquesβ’10 minutes
- Practice activity: Applying PEFTβ’70 minutes
- Walkthrough: Applying PEFT (Optional)β’0 minutes
- Practice activity: Applying LoRAβ’30 minutes
- Walkthrough: Applying LoRA (Optional)β’0 minutes
- Practice activity: Applying QLoRAβ’85 minutes
- Walkthrough: Applying QLoRA (Optional)β’0 minutes
- Practice activity: Comparing fine-tuning techniquesβ’100 minutes
- Evaluating fine-tuned modelsβ’10 minutes
- Detailed explanation of evaluation metricsβ’10 minutes
- Practice Activity: Applying evaluation metrics in fine-tuning modelsβ’65 minutes
- Walkthrough: Applying evaluation metrics in fine-tuning models (Optional)β’0 minutes
- Practice activity: Fine-tuning an LLMβ’65 minutes
13 assignmentsβ’Total 78 minutes
- Graded quiz: Fine-tuning LLMsβ’30 minutes
- Reflection: Setting up your environment in Microsoft Azureβ’3 minutes
- Reflection: Creating your code repositoryβ’3 minutes
- Reflection: LLM fine-tuningβ’3 minutes
- Reflection: Model and dataset selectionβ’3 minutes
- Reflection: Preparing a dataset for fine-tuningβ’3 minutes
- Reflection: Applying PEFTβ’3 minutes
- Reflection: Applying LoRAβ’3 minutes
- Reflection: Applying QLoRAβ’3 minutes
- Reflection: Comparing fine-tuning techniquesβ’3 minutes
- Knowledge check: Fine-tuning techniquesβ’15 minutes
- Reflection: Applying evaluation metrics in fine-tuning modelsβ’3 minutes
- Reflection: Fine-tuning an LLMβ’3 minutes
In this module, you will delve into the critical processes and methodologies involved in fine-tuning LLMs to enhance their performance for specific tasks. By the end of this module, you will have a comprehensive understanding of fine-tuning techniques and be equipped to apply these methods to enhance LLMs for specific, practical applications.
What's included
5 videos13 readings7 assignments
5 videosβ’Total 35 minutes
- Introduction to AI agentsβ’5 minutes
- Differences in multi-agent systemsβ’6 minutes
- Use case demonstration: Multi-agent systemsβ’8 minutes
- Real-world examples: Effective AI troubleshootingβ’6 minutes
- Walkthrough: Designing an intelligent troubleshooting agent (Optional)β’10 minutes
13 readingsβ’Total 345 minutes
- Detailed explanation of principles and architecture of AI agentsβ’10 minutes
- Understanding multi-agent systemsβ’10 minutes
- Principles of multi-agent systemsβ’10 minutes
- Practice activity: Multi-agent systems vs. single agent systemsβ’45 minutes
- Walkthrough: Multi-agent systems vs. single agent systems (Optional)β’0 minutes
- Designing intelligent troubleshooting agentsβ’10 minutes
- Requirements definition for intelligent troubleshooting agentsβ’10 minutes
- Requirements for effective AI troubleshootingβ’10 minutes
- Practice activity: Key requirements for AI troubleshootingβ’85 minutes
- Walkthrough: Key requirements for AI troubleshooting (Optional)β’0 minutes
- Discussion: Best practices for AI troubleshootingβ’60 minutes
- Summary: AI agentsβ’10 minutes
- Practice activity: Designing an intelligent troubleshooting agentβ’85 minutes
7 assignmentsβ’Total 96 minutes
- Graded quiz: AI agentsβ’42 minutes
- Knowledge check: AI agentsβ’15 minutes
- Reflection: Multi-agent systems vs. single agent systemsβ’3 minutes
- Knowledge check: Multi-agent systemsβ’15 minutes
- Knowledge check: Designing intelligent troubleshooting agentsβ’15 minutes
- Reflection: Key requirements for AI troubleshootingβ’3 minutes
- Reflection: Designing an intelligent troubleshooting agentβ’3 minutes
This module provides a comprehensive introduction to integrating natural language processing (NLP) techniques into the development of intelligent troubleshooting agents. You will learn to implement fundamental NLP methods, design effective chatbot interfaces, and apply sentiment analysis to improve user interactions. By the end of this module, you'll have the skills to build and optimize NLP-driven chatbots for troubleshooting, applying foundational text analysis techniques, creating effective user interfaces, and leveraging sentiment analysis to enhance user interactions.
What's included
7 videos10 readings7 assignments
7 videosβ’Total 51 minutes
- Overview of natural language processing (NLP) techniquesβ’7 minutes
- Walkthrough: Developing the chatbot interface (Optional)β’9 minutes
- Use case demonstration: Sentiment analysisβ’5 minutes
- Walkthrough: Implementing sentiment analysis (Optional)β’7 minutes
- Best practices for integrating NLP componentsβ’6 minutes
- Module summary: NLP for troubleshootingβ’7 minutes
- Walkthrough: Implementing NLP for troubleshooting (Optional)β’8 minutes
10 readingsβ’Total 280 minutes
- Detailed explanation: principles and applications of NLPβ’10 minutes
- Developing a chatbot interfaceβ’10 minutes
- Practice activity: Developing the chatbot interfaceβ’30 minutes
- Overview: sentiment analysisβ’10 minutes
- Explanation of sentiment analysisβ’10 minutes
- Practice activity: Implementing sentiment analysisβ’75 minutes
- Integrating NLP componentsβ’20 minutes
- Practice activity: Integrating NLP componentsβ’55 minutes
- Walkthrough: Integrating NLP components (Optional)β’0 minutes
- Practice activity: Implementing NLP for troubleshootingβ’60 minutes
7 assignmentsβ’Total 72 minutes
- Graded quiz: Implementing NLP for troubleshootingβ’30 minutes
- Knowledge check: NLP techniquesβ’15 minutes
- Reflection: Developing the chatbot interfaceβ’3 minutes
- Reflection: Implementing sentiment analysisβ’3 minutes
- Knowledge check: Sentiment analysisβ’15 minutes
- Reflection: Integrating NLP componentsβ’3 minutes
- Reflection: Implementing NLP for troubleshootingβ’3 minutes
This module equips you with the skills to develop a sophisticated troubleshooting agent using Python. The module covers coding core functionalities, integrating ML models, implementing decision-making algorithms, and establishing robust error-handling and logging systems. By the end of this module, you will have a comprehensive understanding of how to build and refine a troubleshooting agent using Python. You will be equipped with skills in coding core functionalities, integrating ML for problem classification, implementing decision-making algorithms, and ensuring robust error handling and logging.
What's included
6 videos19 readings9 assignments
6 videosβ’Total 42 minutes
- Walkthrough: Coding a troubleshooting agent in Python (Optional)β’8 minutes
- Walkthrough: Implementing classification models (Optional)β’8 minutes
- How to implement a decision-making algorithm in Pythonβ’7 minutes
- Walkthrough: Creating a solution recommendation system (Optional)β’7 minutes
- Walkthrough: Implementing logging in ML systems (Optional)β’6 minutes
- Walkthrough: Implementing the troubleshooting agent (Optional)β’5 minutes
19 readingsβ’Total 500 minutes
- Core functionality of a troubleshooting agentβ’10 minutes
- Explanation of key componentsβ’10 minutes
- Practice activity: Coding in Pythonβ’30 minutes
- Problem classification modelsβ’10 minutes
- Explanation of classification modelsβ’10 minutes
- Practice activity: Implementing classification modelsβ’30 minutes
- Practice activity: Implementing and evaluating classification modelsβ’90 minutes
- Walkthrough: Implementing and evaluating classification models (Optional)β’0 minutes
- Decision-making algorithmsβ’10 minutes
- Practice activity: Creating a decision-making algorithm in Pythonβ’30 minutes
- Walkthrough: Creating a decision-making algorithm in Python (Optional)β’0 minutes
- Practice activity: Solution recommendationβ’90 minutes
- Error handling and loggingβ’10 minutes
- Explanation of error handlingβ’10 minutes
- Practice activity: Implementing mechanismsβ’90 minutes
- Walkthrough: Implementing mechanisms (Optional)β’0 minutes
- Practice activity: Loggingβ’30 minutes
- Summary: Troubleshooting agentsβ’10 minutes
- Practice activity: Implementing the troubleshooting agentβ’30 minutes
9 assignmentsβ’Total 54 minutes
- Graded quiz: Troubleshooting agentsβ’30 minutes
- Reflection: Coding in Pythonβ’3 minutes
- Reflection: Implementing classification modelsβ’3 minutes
- Reflection: Implementing and evaluating classification modelsβ’3 minutes
- Reflection: Creating a decision-making algorithm in Pythonβ’3 minutes
- Reflection: Solution recommendationβ’3 minutes
- Reflection: Implementing mechanismsβ’3 minutes
- Reflection: Loggingβ’3 minutes
- Reflection: Implementing the troubleshooting agentβ’3 minutes
This module focuses on the critical aspects of ensuring the quality and performance of troubleshooting agents through rigorous testing, performance monitoring, optimization, and real-world evaluation. You will develop skills to design test cases, implement monitoring systems, enhance response efficiency, and assess the agent's effectiveness in practical applications. By the end of this module, you will have the expertise to rigorously test, monitor, and optimize troubleshooting agents, ensuring they perform effectively and efficiently in real-world situations.
What's included
13 videos8 readings6 assignments1 peer review
13 videosβ’Total 72 minutes
- Designing test casesβ’7 minutes
- Walkthrough: Designing test cases for ML systems (Optional)β’7 minutes
- Hear from an expert: Accounting for cultural, language, and contextual nuancesβ’5 minutes
- Explanation of optimization techniquesβ’7 minutes
- Walkthrough: Implementing optimization techniques (Optional)β’7 minutes
- Walkthrough: Evaluating agent effectiveness (Optional)β’7 minutes
- Hear from an expert: Designing with the end user in mindβ’3 minutes
- Summary: Testing and optimizing the agentβ’6 minutes
- Walkthrough: Testing and optimizing the ML agent (Optional)β’7 minutes
- Hear from an expert: Resolving unexpected issues during implementationβ’5 minutes
- Course summaryβ’4 minutes
- Walkthrough: Producing a troubleshooting agent (Optional)β’4 minutes
- Congratulations on completing the course!β’3 minutes
8 readingsβ’Total 180 minutes
- Explanation of test case designβ’10 minutes
- Practice activity: Designing test cases for ML systemsβ’30 minutes
- Optimizing response time and accuracyβ’10 minutes
- Practice activity: Implementing optimization techniquesβ’30 minutes
- Evaluating agent effectivenessβ’10 minutes
- Practice activity: Evaluating agent effectivenessβ’30 minutes
- Practice activity: Testing and optimizing the agentβ’30 minutes
- Course assignment: Producing a troubleshooting agentβ’30 minutes
6 assignmentsβ’Total 45 minutes
- Graded quiz: Testing and optimizing the agentβ’30 minutes
- Reflection: Designing test cases for ML systemsβ’3 minutes
- Reflection: Implementing optimization techniquesβ’3 minutes
- Reflection: Evaluating agent effectivenessβ’3 minutes
- Reflection: Testing and optimizing the agentβ’3 minutes
- Reflection: Producing a troubleshooting agentβ’3 minutes
1 peer reviewβ’Total 90 minutes
- Course assignment: Drafting the technical reportβ’90 minutes
Earn a career certificate
Add this credential to your LinkedIn profile, resume, or CV. Share it on social media and in your performance review.
Instructor
Offered by
Explore more from Software Development
- Status: Free TrialM
Microsoft
Course
- Status: Free Trial
- Status: Free Trial
Course
- Status: Free TrialC
Coursera
Course
Why people choose Coursera for their career
Learner reviews
- 5 stars
84.21%
- 4 stars
0%
- 3 stars
0%
- 2 stars
0%
- 1 star
15.78%
Showing 3 of 19
Reviewed on Jun 8, 2025
Solid course, really enjoyed. Good work Microsoft, this whole program is quite good.
Frequently asked questions
You should have completed the first two courses in the program, or have equivalent experience with the concepts taught in those courses.
You will need a license to Microsoft Azure (or a free trial version) and appropriate hardware. Note: the free trial version of Azure is time limited and may expire before completion of the program.
To access the course materials, assignments and to earn a Certificate, you will need to purchase the Certificate experience when you enroll in a course. You can try a Free Trial instead, or apply for Financial Aid. The course may offer 'Full Course, No Certificate' instead. This option lets you see all course materials, submit required assessments, and get a final grade. This also means that you will not be able to purchase a Certificate experience.
More questions
Financial aid available,
ΒΉ Some assignments in this course are AI-graded. For these assignments, your data will be used in accordance with Coursera's Privacy Notice.
