I received the title of "PhD in Physics" in March 2020 from University of Roma Tre. I have 6+ years of experience in academia with a focused research in Observational Cosmology. My research projects were aimed at data analysis and estimating statistical quantities that measured the growth rate of cosmic structures like galaxies and clusters of galaxies. I used both simulated and real datasets imported from existing experiments, and also during my PhD thesis, I developed an automated coding pipeline to generate simulated catalogues of galaxies that mimicked the characteristics of the ones that will be observed by upcoming EUCLID space telescope. I also contributed as a co-author in Dolfi et al (2019) where I computed a fit for halo mass function for radio sources from TGSS survey following the estimation of halo bias for specific sources. For computational and programming purposes I speak fluently in C++ and Python with the machines and can manage most of the basic communication in BASH. As of now I am adding SQL (mySQL) to my profeciency in order to efficiently pull data for my projects. To get fluent with Machine Learning frameworks I have been teaching myself with Tensorflow and Keras.

I started working as a Freelance Science Educator during my Bachelor's degree in Physics in Pune. I organised and coordinated many science events being active member of Fergusson College's Astro Club, IUCAA's Public Outreach and a science NGO called "Khagol Vishwa". Most of these events were organised in rural Maharashtra and Goa, aimed at giving best exposure of basics of Astronomy and sparking enthusiasm for Science amongst high-scool kids. During my Masters programme at University of Bonn, I worked as Student Assistant within the graduate school administration as well as for public outreach activities of University of Cologne where I learned basic web-dev and maintenance and creating physics animations using Blender.

When I started with my first job in industry, I got the opportunity to learn, build and deploy speech-to-text models in German and Dutch using KALDI toolkit. I constructed a coding pipeline that takes input as labelled data (spoken utterances with transcripts), prepares it for training, trains the model, deploys the model to the server that takes spoken audio file as input and spits-out the transcripted text. Currently I am building an open-source Marathi corpus that can be used to train and deploy speech-to-text model in Marathi.