Hello! I am

Amogh Desai

An Open-Source Enthusiast Leveraging the Power of Data Science to Create Real-Word Impact

About Me

Passionate Data Scientist based in Mumbai, India


  • 🎓 Senior-year undergrad at Mumbai University
  • đŸ§Ș Applying Data Science and Machine Learning to diverse domains like Astrophysics, Social Science, Agriculture, Education, Medical Science, and Information Management
  • 🏆 Honored as one of the Top 87 Engineers in India by Economic Times
  • 🎯 Identify and solve complex business, operational problems to improve processes
Experience & Projects

Combination of Professional & Research Experience

Experience

  • Research Intern
    (~7.83% acceptance)

    Sasahara Lab, Tokyo Insititute of Technology
    09/21 - Present
    • Mentored by professors and post-doc researchers from the Tokyo Institute of Technology, University of Pennsylvania, University of Aizu and Nagoya University.
    • Collaborating with a team of four, from India, Pakistan, Sri Lanka, Egypt, analyzing Facebook group data to research “The effects of COVID-19 vaccine misinformation on individuals of Asian countries”.
  • Machine Learning Intern

    Vedantu, India’s second-largest ed-tech unicorn
    05/21 - 10/21
    • Implemented a scalable framework for “Detection of multiple questions in user-scanned images” using Natural Language Processing with 85% accuracy, <' /10% FPR & 0.8ms throughput.
    • Tools Used - re, NLTK, OpenCV, pandas, Matplotlib, Seaborn, Scikit-learn, Tensorflow
    • Result - Improved user experience, reduced costs and server load, cleaned 10M+ question database.
  • Student Developer
    (~2.34% acceptance)

    Google Summer of Code’2020, OpenAstronomy
    05/20 - 09/20
    • Mentored by and worked with Dr Matteo Bachetti on stingray, a package used by 40,000+ researchers.
    • Tools Used - NumPy, pandas, SciPy, Dask, Zarr, Numba, CProfiler, memory_profiler, asv, Plotly, Git
    • Result - Enabled analysis of larger-than-memory‌ ‌datasets‌(upto 1Billion points), speed increase for methods.

    Personal Skills

    • Efficient Strategy Planner & Implementer
    • Avid Learner & Logical Thinker
    • Adept Manager & Leader
    • Receptive Listener & Coherent Communicator
    • Proficient in Reporting & Documentation

Projects

  • Helping Hands, Undergraduate Dissertation

    06/21 - 5/22 (Exp.)
    • Developing a gesture-based, brain MRI image manipulation software for surgeons with 3 peers.
    • Uses novel 21-point hand gesture tracking technique, incorporating feedback from radiologists.
    • Tools Used - OpenCV, Scikit-learn, Tensorflow
    • Led a team of 2 peers to develop a multi-lingual web-app to provide crop recommendations to farmers of the state of Maharashtra, India. Considered 30+ factors - soil, weather, market demand etc.
    • Tools Used - NumPy, pandas, NLTK, Plotly, Scikit-learn, Keras, Django, Flask
    • Result - 80% accuracy using deep neural networks.













Technical Skills

  • Data Science

  • Machine Learning

  • Optimization & Parallelization

  • Verion Control & CI

Academia

The Aim of Education is Not Knowledge but Action

Education

  • Bachelor of Engineering (B.E.), Computer Engineering

    Dwarkadas J. Sanghvi College of Engineering, Mumbai, India
    09/19 - 06/22
    • Founding Chairperson of DJS Synapse, department-level machine learning club.
    • Mentee at Unicode, department-level web and mobile app development club.
  • 08/21 - 09/21
    • The curriculum consists of a series of lecture sessions conducted by leading faculty from
      various disciplines, bringing their diverse experiences to the topic of AI.
  • Diploma, Computer Engineering

    Shri Bhagubhai Mafatlal Polytechnic, Mumbai, India
    07/16 - 06/19
    • Elected as Class Representative for 3 years.
    • Winner of State(ISTE-Thakur Poly) and National(IETE-TSEC) Technical Paper Presentation
      Competition for the paper “Various‌‌ Recommendation‌‌ System‌‌ Techniques”.

Publications

  • Desai, A., & Kanani, P. (2021). “Checkmate: A Novel Fact Checker Tool”, submitted; to-be-published.
  • Desai, A., Parekh, V., Unadkat, U., Shekokar, N. (2021). “A comparative study of various asymmetric public key algorithms”, submitted; to-be-published.
Outreach

Small Acts, When Multiplied by Millions of People, Can Transform the World

Open-Source Contributions & Experience

    • Designed functional, performance tests. Converted the project to an object-oriented structure.
    • Tools Used - NumPy, pytest, Astropy, Git
    • Result - 30% increase in throughput, cleaner code to increase usability and contributions.
  • 05/20 - Present
    • Efficient memory management. Parallelized computations. Enabled continuous benchmarking
    • Tools Used - NumPy, pandas, SciPy, Dask, Zarr, Numba, CProfiler, memory_profiler, asv, Plotly, Git
    • Result - Enabled analysis of larger-than-memory‌ ‌datasets‌(upto 1Billion points), speed increase for methods.

LEADERSHIP & OUTREACH

   
Contact Me

I look forward to a successful working relationship in the future.

  • Call Me

    +91 7021004079
  • Email Us

    Send Email
  • Address

    Mumbai, India
Please Fill Required Fields