Yogesh Kumar Singla

Bangalore, India | +91 8872888871 |ersinglayogesh@gmail.com | Github | Linkedin

PROFESSIONAL SUMMARY

Master's graduate in Systems Science and Engineering from University of Ottawa with extensive experience as a Senior Software
Developer at Infosys, specializing in Big Data, Data Engineering, and Data Science. Proven track record in managing large-scale
projects for retail giants like Sweetgreen, bimbo, and Starbucks, including developing ETL pipelines in AWS, database management
with Snowflake, and creating recommendation models for demand prediction. Azure-certified Data Scientist adept in automated
data science platforms, with a knack for crafting end-to-end solutions leveraging advanced ML algorithms.

SKILLS
Languages & Frameworks: Python, Data Science, NLP, Neural Networks, LSTM, Django, AWS, System Design,
Data Structures, REST API’s, ETL, Supervised and Unsupervised Learning.
Databases: SQL, MySQL, MongoDB (NoSQL)
DevOps & Cloud Services: Docker, Kubernetes K8S, Git, SCM, Jenkins
Tools & Technologies: Matplotlib, Scikit-Learn, Pytorch, TensorFlow, Keras, Tableau, MS Excel, LLMs, Langchain.

EXPERIENCE
DATA SCIENCE CONSULTANT DECEMBER 2023 – PRESENT
GENPACT | GOLDMAN SACHS
 Led the development of a machine learning model to address the loan syndicate problem using supervised learning
 Engineered features from diverse data sources including financial records, credit history, and socio-economic indicators to
enhance model accuracy.
 Utilized advanced algorithms like Random Forest, Gradient Boosting, and Neural Networks for precise prediction of loan
defaulters, fine-tuning via hyper parameter optimization.
 Integrated the machine learning solution seamlessly into the existing infrastructure, enabling real-time predictions and
decision-making for risk assessment.

SOFTWARE DEVELOPER INTERN JANUARY 2023 – APRIL 2023

UNIVERSITY OF OTTAWA
 Configuring and deploying servers to manage data and API’s with proficiency in Postgres for designing and implementing
database schemas.
 Implementing scalable architecture using Docker to manage a high volume of requests, optimizing system performance by
conducting testing and optimization of the database.

SPECIALIST PROGRAMMER SEPTEMPBER 2019 – AUGUST 2022

INFOSYS |Bimbo project
 Acquired proficiency in Big Data technologies and advanced fuzzy matching techniques.
 Implemented string-matching algorithms, entity recognition, NLP, and BERT model for robust fuzzy logic applications.
 Leveraged training to enhance data quality and accuracy in large-scale datasets, optimizing information retrieval and analysis
processes.

INFOSYS |SWEETGREEN PROJECT

 Developed a Machine Learning-based recommendation system for product suggestions by leveraging collaborative filtering,
content-based filtering, and deep neural networks.
 Utilized Snowflake for efficient data warehousing, ensuring seamless integration and accessibility of structured and
unstructured data.
 Implemented data pipelines using Azure data services, PySpark, and Apache Airflow to extract, transform, and load data
from diverse sources.

INFOSYS | INFOSYS INTERNAL PROJECT

 Worked on Infosys Data Science and ML Tool to streamline data analysis and modeling for the organization.
 As a Data Scientist, responsible for data preparation including wrangling, cleaning, and preprocessing for modeling.
 Architected and developed AutoML framework, defining hyperparameters and optimization strategies like Bayesian
optimization for model selection and tuning.
  Implemented a ML Flow tool to track model versions and metrics, enabling better management and optimization of
machine learning models deployed on the platform.
 Utilized NLP techniques and BERT model for data preprocessing, cleaning, entity linking, and named entity recognition.

EDUCATION
Masters in System Science and Engineering (Full time, In-person) SEPTEMBER 2022 – December 2023
University of Ottawa (CGPA: 92/100) Canada
B.Tech in Computer Science and Engineering AUGUST 2015 – MAY 2019
Punjabi University (CGPA: 74/100) India

PROJECTS
Udacity Self-Driving Car Engineer Nanodegree Capstone Project
[TensorFlow, OpenCV]
 Implemented perception algorithms utilizing computer vision techniques such as image processing and feature extraction.
 Developed localization modules using sensor fusion techniques including Kalman filters and particle filters.
 Designed path planning algorithms to generate safe and efficient trajectories considering dynamic obstacles and traffic
regulations.
 Integrated control systems for vehicle actuation, including throttle, braking, and steering, to follow planned trajectories
accurately.

GPT Implementation from Scratch with Multi-Head Attention

[Python, PyTorch]
 Implemented the core components of the Transformer architecture, including self-attention mechanisms and feed-forward
neural networks, using Python and deep learning framework PyTorch.
 Incorporated multi-head attention mechanisms to enable the model to focus on different parts of the input sequence
simultaneously, enhancing its ability to capture long-range dependencies and context.
 Fine-tuned the model's hyperparameters and architecture to optimize performance metrics such as perplexity and text
coherence.
 Integrated techniques such as positional encoding and layer normalization to stabilize training and improve model
convergence.

Stock Price Prediction using Multimodal Signals and Advanced Algorithms

[Python, NumPy, scikit-learn, pytorch]
 Curated and preprocessed a comprehensive dataset comprising over 100 financial signals and studies, including technical
indicators, fundamental metrics, and market sentiment features.
 Conducted exploratory data analysis (EDA) and time-series decomposition to identify patterns, trends, and seasonality in
stock price movements.
 Implemented machine learning algorithms such as linear regression, random forests, and gradient boosting machines
(GBM) to model the relationship between input features and future stock prices.
 Developed deep learning models, including recurrent neural networks (RNNs) and long short-term memory networks
(LSTMs), to capture temporal dependencies and nonlinear relationships in the data.
 Evaluated model performance using appropriate metrics such as mean squared error (MSE), mean absolute error (MAE),
and root mean squared error (RMSE), and compared against baseline benchmarks.

CERTIFICATIONS
Certified Azure Data Scientist Associate (Issued by Azure)
Natural Language Processing Nanodegree (Issued by Udacity)
Deep Learning Nanodegree (Issued by Udacity)

AWARDS
Infosys Unit level hackathon (1st/Unit)
3 “Insta Awards” for outstanding contribution in Infosys project (Project level)
Infosys Hackathon organised by ABN Amero (4th/50,000)
Huawei ICT competition (top 150/India)
Chess tournament (3rd/State)