I, Shailesh Kumar, am a part-time researcher, part-time developer, part-time manager and part-time procrastinator.

I have around 20 years of experience in software design, development, support, and management. My areas of interest include digital media technologies, machine learning, computer vision, cloud computing, and SaaS.

I usually spend half of my day answering all the emails, Skype or phone calls, chats, helping people resolve knotty issues, making slides, fixing Excel sheets, or filling up some or other paperwork.

The other half is spent learning something new or building something useful. I still dream that I will be able to build and leave behind something worthwhile before I go.

In my leisure, I dabble a bit in learning about business, finance, history, philosophy and music.


I graduated in 2002 from the department of Electronics and Electrical Communications Engineering, Indian Institute of Technology, Kharagpur, India.

Currently, I am pursuing my part-time Ph.D. from Bharti School of Telecommunications, Indian Institute of Technology, Delhi, India.

Work Experience

I joined Interra Systems right after graduation. At Interra, I have been involved with the development of the flagship products Vega, Baton, Orion, and Winnow. I am also driving Interra Cloud Services SaaS platform.

Currently, I am working as a Director of Engineering at Interra.


I have done coding in a variety of languages: Intel assembly, TI DSP, C, C++, Java, C#, Lisp, Haskell, Rust, Go, MATLAB, R, PHP, Perl, Verilog, etc.

My current workhorses are Python and JavaScript.

It took me a while to get out of the dogma of OOPS. Functional programming probably had the deepest impact on my software design thinking. Haskell is mind-blowing. However, day to day programming works well with multi-paradigm languages. I believe that C++ and Python are decent practical solutions.

What really matters is to choose the right tool for a job based on skill, experience, and fitness for the job. For web development, I like Vue and NodeJS these days and for Deep Learning: TensorFlow and related libraries.

I prefer the Kanban framework for agile software development.

A Salute

What I have learned so far is all thanks to the written works of great scientists and engineers around the world (their books, articles, code).

I would not have been where I am without all the great open-source software available to everyone. I try to contribute and give back as much as I can.

Me on: Twitter | FaceBook | LinkedIn | GitHub | Medium | Stack Overflow | Math Exchange