"we are from different worlds," said Maya to Jay
"So glad we met, Maya," Jay said as he gently pulled the chair for her. She smiled as he took the other chair at the table. Light music...
top of page
Disclaimer: All content posted in my blogs and on my website are my thoughts, and do not represent the opinion of any organization that I work for or have been associated with.
Search
`Do not overdo it.' said the dream merchant
"This feels good. I'm loving this. I Love everyone and everything in this dream." Just as Jay was having his longest dream so far, he was...
A dying man's final goodbye to an aging machine
"Doctor says I have a few months left," said Som as he patted the honing machine, "it's time to leave the world. But I had to come to see...
Build "Purpose" into your product!
The third P of my inbound product management framework (Problem > Persona > Purpose > Product) is quintessential. Most often than not, I...
The 4 Ps of inbound product management
While this series of posts will be themed around ‘productizing’ AI, I will not be surprised if they apply to productizing any idea or...
AI where humans cannot make a decision - Part 2
Sometimes decisions are hard coded using business rules. These business rules get complicated over time. As data, entities and...
AI where humans cannot make decisions - Part 1
I have often been asked for an easy way to identify use cases for AI. Of the many perspectives possible, one is ”use AI where humans...
Prioritize processes and decision points for AI
Once you understand the role of AI in meeting the prime metric(s) and purpose of your organization, the next logical step would be to...
Understand your organization’s primary metric
My previous post was about understand and aligning with organizational purpose before defining your AI use cases. Another thing to...
Understand purpose before use cases
I have seen a tendency in business users to make a list of use cases that could benefit from AI. This typically happens because of...
Planning AI Use cases
This category of blog posts will focus on planning AI use cases for your organization. As you read through multiple use cases and...
Prediction end point - streaming or batch?
We have discussed a few aspects of defining a prediction end point. Here is one more. Always work with your business users to understand...
Prediction end point - how comprehensive should it be?
While defining the prediction rule end point, I have often seen the spec designed such that the data science behind it becomes logic...
Defining prediction API - some tips
The end result of a requirement gathering exercise for AI is often the end point definition for the desired prediction rule. Defining the...
Setting minimum acceptance level for AI
A question I have been asked often, “Should we aim for 70% or 80% or 90% accuracy with our prediction rule?” Like most questions in life,...
Pilot your prediction rules before go live
So you have the prediction rule available. The model has been trained, and you have reasonable amount of confidence in it’s estimated...
The need to stress test AI
Often AI is a high compute environment. At prediction time, the need for computational horsepower might be high, specially for...
Test AI with known data
So the data science is done, programming completed and the prediction API ready. It might also give you an estimate of expected accuracy...
Meaningful training dataset
Supervised learning needs a training dataset to learn from. It is important to understand the nuances though. A meaningful training...
Plan on measuring success of AI
So how will you measure success of AI? As you start planning your AI implementation, ensure that there is enough thought given to it’s...
bottom of page