SY

Hi, I'm

Sourav Yadav

PhD Researcher, Computer Science — University of Calgary

I research trustworthy, decentralized computing — blockchain, digital identity, and distributed ML — applied to domains like healthcare, smart grid, agriculture, and supply chain. Before starting my PhD, I spent 8+ years in industry as a data scientist and ML engineer, shipping federated learning systems into production and publishing peer-reviewed research along the way.

01 About

I'm a PhD researcher in Computer Science at the University of Calgary, having started in August 2026 after 8+ years of industry experience across data analysis, machine learning engineering, and applied research — most recently as a Senior Data Scientist at Orion Innovation. My research centers on trustworthy computing and distributed technologies — developing reliable, secure solutions built on decentralized and distributed technologies such as blockchain, digital identities, and distributed ML. I'm particularly interested in the convergence of blockchain with domains like smart grid, healthcare, agriculture, and supply chain. I bring an industry grounding in shipping federated learning and ML systems into production, built on scalable pipelines with Spark and Kubernetes, into this academic research on decentralized, trustworthy digital ecosystems.

  • 8+Years in industry
  • 5Companies
  • 3Publications
  • 11Certifications

02 Skills

Languages

PythonShell ScriptingSQLMySQL

Data Science & ML

NumPyPandasSciPyTensorFlow PyTorchScikit-learnXGBoostOpenCV SparkPySparkHDFS

Generative AI & LLMs

LangChainOpenAI GPTHugging Face TransformersPGVectorPinecone

ML Theory & Practice

Probability TheoryHypothesis TestingA/B Testing Statistical ModellingPredictive AnalyticsAutoML Hyperparameter OptimizationModel DeploymentExplainability

Databases

MySQLPostgreSQLMongoDBNeo4jSnowflake

Big Data

HadoopYARNSparkPySpark EMRDatabricks

Cloud & DevOps

AWSGCPAzureMLDocker KubernetesCI/CDTerraformGitFastAPI

Visualization

TableauGoogle Data StudioStreamlit

Specialized Domains

Federated LearningDifferential PrivacySecure Aggregation Healthcare Data / EHRTime Series ForecastingComputer Vision Risk & Financial ModellingDistributed Systems

03 Education

PhD, Computer Science

University of Calgary · Calgary, AB, Canada

Aug 2026 – Present

Master of Science, Computer Science

Illinois Institute of Technology · Chicago, IL

Aug 2019 – May 2021

B.Tech, Electronics and Communication Engineering

West Bengal University of Technology · Kolkata, India

Aug 2011 – May 2015

Certifications

04 Experience

Sr. Data Scientist

May 2025 – Jul 2026

Orion Innovation Inc · Remote

  • Led the design and implementation of decentralized ML frameworks for healthcare applications, focusing on privacy-preserving techniques.
  • Developed and optimized classical machine learning models to predict patient outcomes using regression and tree-based approaches.
  • Implemented deep learning models using TensorFlow for medical image analysis, achieving high accuracy in diagnostics.
  • Collaborated with cross-functional teams to integrate ML solutions into production systems using REST APIs.
  • Conducted research on federated learning algorithms, publishing findings in industry conferences.
  • Utilized Kubernetes for scalable ML pipeline deployments, ensuring efficient resource management.
  • Mentored junior data scientists on best practices in deep learning and decentralized optimization techniques.

Data Scientist / Machine Learning Engineer

Jan 2022 – Oct 2024

Health Solutions Research Inc · Chicago, IL

  • Developed federated learning models to train ML systems across sensitive healthcare data while ensuring patient privacy.
  • Implemented secure aggregation and differential privacy strategies to enhance data security in ML applications.
  • Engineered and fine-tuned deep learning models using PyTorch for clinical data prediction and anomaly detection.
  • Designed scalable ML pipelines utilizing distributed systems like Spark, improving data processing times significantly.
  • Collaborated with healthcare professionals to ensure model relevance and clinical validity.
  • Researched the integration of Large Language Models into healthcare systems for patient interaction and decision support.
  • Published results of innovative research in decentralized learning at leading AI conferences.

Data Scientist

Feb 2021 – Dec 2021

Spring ML Inc · Chicago, IL

  • Developed ML models for various clients, focusing on optimization and performance tuning.
  • Built deep learning applications for real-time data processing using Python and TensorFlow.
  • Participated in the design of data pipelines to streamline data collection and processing.
  • Contributed to the deployment of ML models in cloud environments for scalability and reliability.
  • Provided training sessions for team members on advanced ML concepts and tools.

Data Scientist

Jun 2017 – Jul 2019

Wipro Technologies · Bengaluru, India

  • Developed predictive models for clients in healthcare and finance, improving operational efficiency.
  • Implemented machine learning algorithms for data analysis and predictive modeling, resulting in actionable insights.
  • Collaborated with cross-functional teams to integrate ML solutions into client systems.
  • Deployed and managed ML models on AWS.
  • Conducted training on machine learning techniques for junior staff.

Data Analyst

Nov 2015 – May 2017

Wipro Technologies · Bengaluru, India

  • Performed data cleaning and preprocessing to prepare datasets for machine learning applications.
  • Developed dashboards and reports to provide insights into data trends and patterns.
  • Assisted in developing ML models through data analysis and feature engineering support.
  • Worked with SQL and Python to manipulate large datasets effectively.
  • Presented findings to management, contributing to data-driven decision-making.

05 Projects

Jan 2020

Scalable Meta-Learning for Few-Shot Classification

Designed a meta-learning framework leveraging episodic training and prototypical networks, achieving state-of-the-art performance on benchmark few-shot learning datasets with strong adaptability to unseen classes.

PythonPyTorchNumPy scikit-learnMatplotlib
Aug 2020

Interpretable Deep Learning for High-Stakes Applications

Investigated interpretability methods for deep learning in critical domains like healthcare and finance; implemented a model-agnostic explanation framework using SHAP and LIME for transparent, reliable decision-making.

PythonTensorFlowKeras SHAPLIMEMatplotlib

06 Publications & Conferences

07 Contact

Open to conversations about trustworthy computing, blockchain, distributed ML, and research collaborations. Reach out any time.