2. ABOUT ME
DINESH SHARMA
Delivery Head
Over 26 years of experience in IT Industry with more than 15
years in Project management. Extensive Project, Program and
Delivery Management and more than 10 years into Agile Based
Models.
Worked with a number of organizations both product based and
service based across the globe. Experience working with wide
range of domains and technologies. Exceptional track record of
delivering a high number of projects and programs with 100%
success rate. Authored and published a large number of articles
and whitepapers on many topics in Agile and Project
Management.
Wunderman Thompson
Commerce and Technology
(A WPP Company)
3. www.agilemumbai.com
1. Introduction
2. Power of AI
3. Agile Project Management
4. A correlation
a). Applying AI in Agile Project
Management
b). Applying Project Management in AI
projects.
5. Conclusion
6. Q & A
AGENDA
4. www.agilemumbai.com
INTRODUCTION
What is AI?
• AI is a collection of different technologies that can be brought together to
enable a machine to act with intelligence - to learn and make decisions.
• Artificial Intelligence (AI) has started making a substantial impact to many
parts of our society, it has already made its place in our lives.
• A true AI system is one that can get smarter and more aware through
learning, allowing it to enhance its capabilities and its knowledge over time.
What is Agile?
Agile methods (e.g. Scrum) have been widely used in industry to manage
software projects.
“Integrating AI with Agile
methodologies creates a
dynamic duo,
revolutionizing project
management through
enhanced efficiency
leading to lesser impact on
environment and better
business sustainability”
In this presentation, we will be discussing how we can bring Artificial Intelligence and Agile together - Focusing on
different co-relations.
5. www.agilemumbai.com
POWER OF AI
Process large amounts of data
Automate repetitive, high-volume tasks
Analyze large amounts of data to
identify patterns
Make autonomous decisions
Providing actionable recommendations
& significantly reduce errors
Analyze new information, learn from
data and results in near real time
BENEFITS
Enables project analytics and
insights for scheduling, progress
tracking estimation and risk
prediction and many more areas
Augmented
Helps to accelerate productivity,
increase accuracy and precision
and thus helps project success
rates.
Assisted
AI can learn from experience,
adjust to new inputs, and perform
human-like tasks.
Autonomous
6. www.agilemumbai.com
AGILE PROJECT MANAGEMENT
• Agile Project Management is an approach based on delivering requirements iteratively and incrementally throughout the project life cycle.
• Agile reduces difficulty of planning the project, executing the product and then testing the product for flaws by breaking these big cycle into
small cycle or segments.
• The Agile Manifesto in Project Management is a formal proclamation of 4 Key values and 12 principles to guide an iterative and people-
centric approach to software development.
• Scrum is one of the most common Agile Frameworks and sometimes even used as a synonym of agile.
• The rapid development of artificial intelligence (AI) poses new opportunities and challenges for Agile project management.
We all know
What Agile is ?
"Agile is not a
methodology, it's a
mindset of
continuous
improvement."
8. • Scrum, the most popular of team-level Agile frameworks, known for its versatility.
• Scrum is all about iterative progress, quick adaptability, and delivering incremental value.
• Its structure is based on defined roles (Scrum Master, Product Owner, Development Team), events
(Sprint Planning, Daily Scrum, Sprint Review, Sprint Retrospective), and artifacts (Product Backlog,
Sprint Backlog).
“Imagine enhancing your Scrum practices using the Power of AI”
APPLYING AI IN AGILE SCRUM
9. The integration of Scrum and AI Tools
“It holds the potential to revolutionize how we navigate Scrum practices”
APPLYING AI IN AGILE SCRUM
SPRINT PLANNING
• ChatGPT prompt can help you sketch out an
effective strategy and prioritize tasks for your
software development project.
• Spinach io can document your sprint goals, create
missing tickets, update existing tickets, suggest
estimates based on previous work completed,
prioritize items in the backlog based on the sprint
goal, and even predict potential blockers.
DAILY STAND-UP
• ChatGPT can help structure these crucial meetings to
maximize their value.
• Spinach io integrates with your current tools like
Slack, Teams, Jira providing context from the prior
standup, keeping the discussion focused, sharing a
standup summary, and suggesting follow-up actions
• Stepsize io can help you generate summary on
team's progress, blockers with data driven insights.
SPRINT RETROSPECTIVE
• Integrating big data analytics into retrospective.
• AI can compile key metrics and insights, generate
reports, compare it historically and identify areas of
improvement.
• Quality assurance analysis
• Velocity analysis
SPRINT REVIEW
• AI-driven applications can automate the generation
of presentation slides for sprint review meetings.
• Analyze completed sprint issues and creates slides
based on issue content.
• A powerful combination of OpenAI for content
analysis, integration with Jira's API for real-time data,
cloud infrastructure for scalability, and deep learning
models (like DALL-E) for image generation.
www.agilemumbai.com
10. BENEFITS OF AI IN AGILE SCRUM
Meeting summaries appear magically,
stakeholder recaps are generated with a click,
and ticket updates are suggested without you
even asking.
By automating tasks like creating/updating
tickets and generating meeting notes, AI frees
up time and mental energy for team members
to focus on more creative and complex tasks.
Analyze data from various sources in real time.
AI can identify patterns and trends that even
the most perceptive human Scrum Master
could miss.
By making decisions based on data and metrics,
an AI Scrum Master ensures that your team's
decision-making process remains consistent,
fair, and unbiased.
Automating Repetitive Tasks Real time Data Analysis
Improved Team Productivity
Consistent Decision Making
1
4
3
2
www.agilemumbai.com
11. APPLYING AI IN SCALED AGILE
Continuous Integration/
Continuous Deployment
(CI/CD)
Log Analysis: Implementing AI to scan logs and identify any patterns that could suggest a potential issue.
Rollback Decisions: Based on real-time monitoring and AI analysis, systems can decide when to roll back
a deployment automatically.
Program Increment (PI)
Planning
Capacity Planning: By analyzing past velocity and capacity data, AI can make recommendations for
future PI planning.
Risk Analysis: Use AI to identify potential risks based on historical data and trends.
Automated Testing
Defect Prediction: AI can analyze code to predict where defects might arise, allowing for more targeted
testing.
Flaky Test Detection: Detect and diagnose flaky automated tests that intermittently fail.
Backlog Prioritization &
Management
Prediction Analysis: Use AI to analyze the history of backlog items to predict which ones have a higher
chance of introducing bugs or delays.
Effort Estimation: By analyzing past tasks, AI can suggest the likely story points or man-hours required
for new tasks.
Some ways to integrate AI into the SAFe Framework
www.agilemumbai.com
12. APPLYING AI IN PROJECT MGMT
• Project Charter
• Initiation
• Develop & complete
deliverables
• Status & Tracking
• Update Project schedule
• Objectives
• Quality deliverables
• Effort & Cost Tracking
• Performance
• Postmortem
• Project punch list
• Final reporting
• Scope & Budget
• Create WBS
• Communication plan
• Risk Management
Project
Initiation
Project
Closure
Project
Monitoring &
Control
Project
Execution
Project
Planning
• Analyze historical data,
productivity rates, time
estimates, working hours
for patterns.
• Automate repetitive tasks.
• Identify risks for the
projects
(ChatGPT, Bard, Claude)
• Plan your calendar
(motion)
• Assist in task selection,
keeping track of the tasks
completed, and send
created reports to all
stakeholders (PMotto.ai)
• Chatbots can monitor
changes made to the
source code and report
bugs in any code line
(ChatGPT, Bard, Claude)
• Enable the development
of new ideas and help
accomplish bigger tasks.
• Make recommendation
for complex decision-
making.
(ChatGPT, Bard, Claude)
• Predict cost based on
project size, contract type.
Alerts potential delays,
point out the under perf
based on KPIs
• Recommend ways to bring
the project back on track.
(ChatGPT, Bard, Claude)
• Identify MOMs using
(Zapier AI, Whisper V3)
• Better selection &
prioritization through
predictive analytics.
• AI delves into the history
of past projects and gives
real-time information on
resource engagement
(ChatGPT, Bard, Claude)
• Organizational Knowledge
repository(ChatPDF )
WITH AI POWERED SYSTEM
www.agilemumbai.com
13. APPLYING AI IN PROJECT MGMT
Create your own GPT Model
Train the Model
Deploy the Model
Create and configure the
Model
Organization
PM GPT
Get the Realtime information
from the model
Carry out complex analytics on
the information
Seek Guidance on a specific
project
Initiation Planning Execution Monitoring & Control Closure
Feed the Model and Use the Model
www.agilemumbai.com
14. APPLYING AI IN PROJECT MGMT
Case Study: Creating our own GPT - how did it help us
GPT Created at the organization level and implemented the same.
• All Project Data supplied, with regular updates
• Training provided through interactions
• Additional information with misc documents
• Substantial improvement observed on CPI, Budget Performance
• Marginal improvement in SPI, Schedule Performance
• Immediate improvement in RPI, Scope Performance
• Significant improvement in DDD, Quality Performance
www.agilemumbai.com
15. Improved Project Planning
Better Decision Making
Enhanced Efficiency
Improved Risk Management
Predictive Analytics
Improved Communication & Collaboration
Real-time Project Monitoring & Alerts
ACTION PLAN
CHALLENGES
Integration of AI
Data Privacy and Security Issues
Overcoming Resistance to Change
Technical Limitations
Need for Continuous Improvement
Training and Support
Addressing Cost considerations
Benefits and Challenges
APPLYING AI IN PROJECT MGMT
www.agilemumbai.com
17. Data Dependencies
Team Skillset
Project Timeline
Iteration Impact
Monitoring & Maintenance
Limited reliance on specific data
Software development expertise
Relatively Predictable timelines
Smaller Impact of Scope changes
Stable post-launch maintenance
Heavy dependency on quality data
Data science, ML algorithms, coding
Variable due to data challenges
Scope changes can disrupt Models
Ongoing monitoring, retraining
Scope & Complexity
Requirements
Aspects AI Projects
Traditional IT Projects
Evolving Scope & experimentation
Well defined scope & requirements
Stable & often static Dynamic and subject to Change
APPLYING PROJECT MGMT IN AI
18. Characteristics
• Usually far more
complex, expensive, and
multi-disciplinary than
traditional software
development.
• Getting volumes of data
is a challenge and data
itself is expensive.
• AI/ML expertise are in
scarcity in the industry,
as technology is new.
• Operate on a high level
of innovations where a
lot of factors, and end
results are unknown.
Scrum is a more disciplined, prescribed approach to deal with complexity and uncertainty.
• Teams typically work in sprints, which are short, iterative time periods.
• At the end of each sprint, the team should be able to demonstrate potentially shippable work.
• However, it is often necessary to deploy a successful ML model over multiple sprints. As ML models
can be complex and require a lot of testing and fine-tuning.
• Customers and stakeholders should be involved in the development process to ensure that the ML
model meets their needs.
• They should also be involved in the testing process to provide feedback on the model's performance.
We are also in the discovery path and in every retrospective, we are becoming better to execute AI projects in better ways.
CREATE TRAIN DEPLOY
Gather right data
Problem Statement
Choose right Algorithm
Test the Model
Train the model Deploy the Model
Monitor the Model
4 weeks Sprint
Steps for successful AI execution
APPLYING PROJECT MGMT IN AI
Feedback Loop
www.agilemumbai.com
19. CONCLUSION
AI and Agile mutually enhance each other, fostering more efficient, collaborative and
effective project management.
Future-Forward Strategy
Let’s embrace AI within Agile frameworks to navigate and lead towards more efficient
sustainable business transformations.
www.agilemumbai.com