Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
98 lines (87 sloc) 6.07 KB
# My Resume
## Personal Details:
> Name: Pinaki Nath Chowdhury
> Current Job Title: Project Linked Person (aka. Research Assistant)
> Institution/Organization: Indian Statistical Institute, Kolkata
## Contact Details:
> Email Id:
> Phone (M): +91 7432814640
> Website:
## My current work
> I work in Computer Vision where my research interest are mainly text detection and text recognition,
in the wild and handwritten. With help of my guides, Prof. Umapada Pal and Prof. P. Shivakumara,
I was able to solve some problems and hoping to solve some more. (Shall update more about this later).
## Institutes/Organizations that I have previously worked with:
> Central Drug Research Institute, Lucknow (a CSIR Lab)
>> During my time here, I worked with Dr. Sukant Khurana helping in
crawling information from various News websites in India and South-EAST Asia
and understanding awareness of various diseases and several other social aspects.
> Oxygen Technologies Sdn Bhd, Malaysia
>> Oxygen Technologies is a startup in Malaysia and I was involved in building a POC
where hospital forms could be scanned and understood by a computer, thereby reducing
time and also storing information digitally.
> MyAnatomy Ventures, Bangalore
>> MyAnatomy is a startup in bangalore aiming to provide placement opportunities to all
sudents in Indian colleges. I was the Full Stack Developer for MyAnatomy (MERN stack).
During my stay, I fixed several bugs, developed 2 micro-services; 1 for payment and another
for a new services to customers. Though I was a Full Stack Developer, I mainly developed
the backend, optimized the code to reduce latency etc. In frontend, I only fixed bugs
which were notorious and difficult to debug.
> EduGorilla Community Pvt. Ltd, Lucknow
>> EduGorilla is a EdTech startup originally founded in Lucknow. I worked here during 2016
for nearly 6 months. During my stay here, I would crawl large amounts of information,
optimize the code which used to validate the large information that we had and figure out ways
to reduce latency by using multi-threads (as there were some APIs involved in the process).
Today EduGorilla is one of the most successful EdTechs in India (thanks to Mr. Mangalik)
## Some organizations where I worked for short time:
> Egnify Technologies, Hyderabad
>> Egnify is a startup in Hyderabad whose main goal is how to help school students to score more
by providing a detailed analysis of their performance and where they are lacking
(i.e. Learning Analytics). Egnify has won several awards and has been in news quite a number
of times. I was hired as a Full Stack Developer where my work was primarily to develop their
QMS but during my short stay all I could do is use GraphQL to connect the various microservices
and also the base structure of QMS (but I did not develop QMS).
> Technolosol, Bangladesh
>> While in 3rd year semester break, I did an Android course where I met Mr. Arnab Charkaborty who later asked me
to help build an application to track service cars in Bangladesh and send the information to an ERP.
I helped in the planning stage, providing solutions and also agreed to lead a team of young
developers to build the android application.
## My Bachelor Thesis
> Shapes can tell us a lot of intresting properties of Nature. Just think, what would be the shape
of a water bubble in outer space? (it is spherical. But why?)
My goal was NOT to answer why (that I left to the community :p), I wanted to know "which is the most
common shape in this Universe?"
>> Now, I couldn't solve this problem totally, but came up with some tricks to get an experimental
proof of my idea. I wrote a crawller to download some x-ray images of supernova from places like
Chandra X-Ray Observatory etc. We knew that our small collected data has more number of Type-1
Supernova. (We just needed our algorithms to understand it). We developed a variant of K-Means
algorithm (basically using K-Means to understand the shape based on temperature of Gas Clouds).
Invented an idea of Similarity Matrix (but humans had developed this concept long ago). Then we
needed to represent this information present in high dimension into 2D plane for humans to "see" it.
So we used embedding techniques like PCA, t-SNE and hierarchical clustering for visualisation.
Luckily, our algorithm did say that Type-1 Supernova was the most common one (phew!!).
## My Education
>> St. Xavier's School (Durgapur) [School]
>> D.A.V. Modern School (Durgapur) [High Scool]
>> Kalyani Government Engineering College, Kalyani (not far from Durgapur) [Undergrad College]
## Some small achievements
>> WBJEE (West Bengal Joint Entrance Examination) 2014 Rank: 960 (out of 100,000)
for admission to Engineering Colleges in West Bengal
>> CodeChef Snackdown 2017 Rank: 729 (Got eliminated in the quater Finals)
>> Google CodeJam 2018 Rank: 6367 (every year I solve 3 to 3.5 problems in Qualifying Rounds only to
get eliminated in the next)
## Some little Technical Skills:
>> Programming Language: Python(2&3), JavaScript, C/C++, Shell, Octave/Matlab
>> Frameworks/Library: Tensorflow, PyTorch, Keras, Scikit-Learn, NodeJS, ReactJS, Redux, Scrappy, GraphQL.
>> Database: MongoDB, Google Datastore, MySQL
## References
available on request
You can’t perform that action at this time.