Work History
Education
Skills
Affiliations
Accomplishments
Languages
Timeline
Generic
Prashant Singh

Prashant Singh

Noida

Work History

Senior Machine Learning Engineer

CHILD GROWTH MONITOR (WELTHUNGERHILFE)
09.2022 - 11.2023
  • Successfully developed and deployed a multimodal based deep learning models for measuring child height, improving prediction accuracy from 65% to 72%
  • Conducted comprehensive feature analysis to identify key factors correlated with the height measurement algorithm, elevating overall accuracy from 60% to 65%
  • Optimized the dataset creation pipeline, significantly reducing the time required from 6 hrs to 2 hrs, enhancing efficiency and productivity
  • Collaborated closely with backend and frontend teams to develop a comprehensive dashboard, enabling stakeholders to monitor various business and technical KPIs effectively.
  • Implemented scalable machine learning platforms that supported large-scale deployments of AI-driven applications.

Data Scientist

CHILD GROWTH MONITOR (WELTHUNGERHILFE)
03.2020 - 10.2021
  • Design and Implementation of pointcloud and depthmap based deep learning models to predict height and weight of children
  • Responsible for developing and testing MLOps pipelines to collect, train, evaluate, and deploy Deep Learning models
  • Developing pipelines to preprocess incoming data for inference and prepare datasets in multiple formats for training and evaluating different types of models
  • Analyzing and identifying key features on millions of data points using explanatory analysis to increase the model's accuracy and reliability
  • Implemented an API for deploying GRADCam algorithm to identify any bias in results, and improve the model's performance.
  • Enhanced data-driven decision making by implementing machine learning models and algorithms.
  • Optimized business processes for increased efficiency through detailed data analysis and visualization tools.

Data Scientist

DERMASCREEN UG
11.2018 - 05.2019
  • Research and development of Deep Learning models centered around classifying the skin cancer using RGB images
  • Responsible for increasing accuracy of skin cancer detecting algorithms by 35% on our holdout testset and weighted accuracy of 32% on ISIC public testset
  • Worked on creating infrastructure on Amazon Web services for model training and metrics evaluation.

Machine Learning Intern

UNTERNEHMERTUM
06.2018 - 08.2018
  • Training models to identify drivers licence and extract the information present in license
  • Developing image augmentation pipelines to prepare synthetic dataset to assist in model training and evaluation without data acquisition cost
  • Research and development of object detection algorithm to identify license plates of the car with 94% accuracy.

Work Student

TERRALOUPE GMBH
09.2016 - 09.2017
  • Developed deep learning models to identify the traffic signs on the roads, achieve accuracy of 83%.
  • Applied semantic segmentation and geo-spatial algorithms for calculating the available roof area for installing solar panels across different German cities
  • Implement API to protect multiple logging by same account on the website account.

Software Developer

HCL TECHNOLOGIES LTD.
07.2011 - 08.2014
  • Developed webservices that allows user to place bets using kiosk machines in casino
  • Implemented API's for monitoring messages on EMS queues to prevent server outages.

Education

MSc.in Informatics -

TECHNICAL UNIVERSITY OF MUNICH
05.2018

B.Tech in Computer -

SRM UNIVERSITY
05.2011

Skills

  • Deep Learning
  • Machine Learning
  • Computer Vision
  • LLMs
  • MLOps
  • Tensorflow
  • Pytorch
  • AWS
  • Microsoft Azure
  • Python

Affiliations

  • ChatGPT Prompt Engineering for Developers
  • Building Applications with Vector Databases
  • Machine Learning Modeling Pipelines in Production
  • Deploying Machine Learning Models in Production
  • Finetuning Large Language Models

Accomplishments

    [1] G. S. Bindra, P. K. Singh, K. K. Kandwal, and S. Khanna. Cloud security: analysis
    and risk management of vm images. In 2012 IEEE International Conference on
    Information and Automation, pages 646–651. IEEE, 2012.


    [2] A. Trivedi, M. Jain, N. K. Gupta, M. Hinsche, P. Singh, M. Matiaschek, T. Behrens,
    M. Militeri, C. Birge, S. Kaushik, et al. Height estimation of children under five years
    using depth images. In 2021 43rd Annual International Conference of the IEEE En-
    gineering in Medicine & Biology Society (EMBC), pages 3886–3889. IEEE, 2021.

Languages

English
Bilingual or Proficient (C2)
Hindi
Bilingual or Proficient (C2)
German
Elementary (A2)

Timeline

Senior Machine Learning Engineer

CHILD GROWTH MONITOR (WELTHUNGERHILFE)
09.2022 - 11.2023

Data Scientist

CHILD GROWTH MONITOR (WELTHUNGERHILFE)
03.2020 - 10.2021

Data Scientist

DERMASCREEN UG
11.2018 - 05.2019

Machine Learning Intern

UNTERNEHMERTUM
06.2018 - 08.2018

Work Student

TERRALOUPE GMBH
09.2016 - 09.2017

Software Developer

HCL TECHNOLOGIES LTD.
07.2011 - 08.2014

MSc.in Informatics -

TECHNICAL UNIVERSITY OF MUNICH

B.Tech in Computer -

SRM UNIVERSITY

Similar Profiles

CREATE PROFILE
Prashant Singh