Learning

Good decisions lead to better results, but how to make decision?. What factors you consider when you make decision. The book about Customers are difficult to understand Our intuition is wrong Structured analysis of the data helps us make the right decision Ensuring high data quality is everybody's responsibility  

For the last 3 years I have been working on shipping a large distributed system where many subsystems are powered by machine learned models. Before this I used to work in building cloud services, developing operating system protocols and services. Here are some lessons from this transition so far. Feature begins after you ship the code: Earlier feature design and development stage is where majority of the energy is spent. You write your functional specification, do detailed design , implement the feature, unit test, functional test, end2 end test, stress test and after the feature passes release criterion you release the feature. In machine learned system you need to all of the above, but that's just stage 1 of the feature. Many of these features to work well need large amount of data, for your v1 feature you will not have all the data you need, feature is not yet available in real world so you cannot fully anticipate what will be the real usage of your feature is. 

While watching a movie recently I slowly imagined myself in a role and thinking about what actually I would do. I watched the movie so intensely and could really feel the actor and how he is thinking and acting each moment so vividly.   Is this really possible with a technology.  With hololens, it should be. Let's wait and see, this opens up possibilities.   Experience the movie as a or any character in movie ?. Imagine that!, nothing costs to think about the fun you would have.

Dan is feeling the heat, he has been feeling it for a while on his eye lids. Heat is increasing as time is passing, he needs to decide now, now or never to open the eyes and wake up or find a way to avoid the heat. Finally waking up won the battle of wills and he woke up. Bright gory sun is staring at him through the wall of windows. Now he remembers that he forgot to close the blinds before he went to around 8 pm. Its 10:30 pm whispered angel, that tiny projection in his ear. That tiny projection is where his omniscience personal agent angel lives. Hmm so I only slept 2 and ½ hours he thought. I need to figure how to get my blinds to automatically close he thought. Now you might be thinking (or not) there is something odd about the times mentioned in my introduction above. If you did not catch its ok, hint: “Sun is shining brightly at 10:30 pm?”.   Sun can shine brightly at 10:30 PM, yes it’s certainly in the realm of possibility. That is the main story I wanted to tell , s...

Friday, February 21, 2014 9:58 AM Release criterion plays a critical role in releasing software. Good release criterion includes setting up the right quality bars that product has to pass before it's deemed ready. Traditional software testing methods focused on various test design approaches on coming up with the right set of test cases that assures some measure of quality. How does this work in a world where the software is largely machine learned, how does one come up with a good release criterion.? When shipping initial set of models where models are only trained with limited set of real world data it's hard to estimate how well they do once the models are live and taking the real traffic.

Seattle Half marathon, here I come. I signed up for seattle 2013 half marathon   which will take place on Dec 1, 2013. My current training plan is to run 2.5 miles every other day around the neighborhood. I am using mapmyrun app to track my workouts. Based on todays run I am doing an average of 11 minutes per mile and that puts me around 134 minutes at top pace and a reasonable 150 minutes for a half marathon. Over the next few months my goal is to Keep running regularly. Find different routes on good sunny days. Go up to 5 miles per run. Improve rhythm and running style.  

Often students are told that they must think "critically" if they are to succeed. Can we identify the skills and underlying techniques that allow us to think better? Smart thinking means knowing how to Work out and express your main ideas Plan your communication of ideas so that they can be clearly understood Check to see if you have covered all the important part of your topic Establish a framework or structure in which your basic facts and evidence make sense Present ideas by linking them together to convince readers of your conclusion