Resume

Work
Experience

Aug 2019 - Dec 2021

Led a 3-person team to develop an AI spend optimization app with Power BI and Azure ML, utilizing tree-based ML models (Decision Tree, XGBoost) for feature extraction and clustering algorithms (k-means, k-prototype, hierarchical) for data pattern recognition. Achieved a 40% reduction in retail client expenditure, saving US$ 5M by predicting savings and identifying expense drivers.
Streamlined project workflows by creating advanced big data pipelines with Azure MLOps, MongoDB, and Azure Data Factory.

Led a team of 4 to develop an AI job profiler using GPT-2 for feature engineering and HuggingFace, BERT Transformers, and GANs for text summarization, entity recognition, and job description processing. Standardized job catalogs, reducing HR interview and resume screening by 90%, enhancing efficiency.
Wrote SQL queries on PostgreSQL for ETL pipeline.
Implemented Docker for cross-platform compatibility, managed Kubernetes for scalable API service deployment, and integrated CI/CD pipelines for efficient project updates.

Aug 2018 - Jul 2019

Developed a Tableau credit risk dashboard with XGBoost and statistical algorithms to identify potential loan defaulters for a private lender, assigning credit scores. Achieved a 65% reduction in default losses, saving US$ 3.1M.
Managed a team of 2 to develop a customer segmentation dashboard for an Indian bank, using unsupervised learning on customer data. Enabled targeted campaigns for online stock trading, cutting marketing costs by 50% and boosting acquisition by 30%.
Built and deployed data pipelines on AWS EC2 with Docker containerization.

Aug 2016 - Jul 2018

Developed an AI web app with AWS Sagemaker for sentiment analysis (LSTM) and restaurant review categorization (Ensemble modeling: Gradient Boost, Logistic Regression, Multinomial Naive Bayes) into aspects like food quality, ambience, etc. Used Elasticsearch and Logstash for data processing, cutting review analysis time for the client by 85%.
Led a 3-person team to develop a full-stack Apache Spark application, summarizing U.S. Court Case stories with LSTM. Implemented multi-label classification (Label Powerset) to predict Acts and Laws, reducing case analysis time for the law-firm by 30%.
Managed AWS S3 and SQL server, automating daily updates of machine learning models and data via cron jobs.
Created and deployed Docker images on AWS EC2 instances. Utilized AWS Sagemaker for a secure model development lifecycle, from data acquisition to deployment, following CI/CD practices.
Designed the project with Apache Spark, Elasticsearch, Logstash, Kibana, and Hive for seamless operations.

2022 - 2023

Master of Professional Studies in Analytics, Applied Machine Intelligence

2012 - 2016

Bachelor of Engineering, Electronics and Communication

Techniques: Data Analysis, A/B Testing, Machine Learning, Problem Solving, Deep Learning, Generative AI, Designing RAG Architecture, Natural Language Processing, Feature Engineering, Data Cleaning, Web Crawling, Word Embeddings, Backend Development, CI/CD pipelines, Data Visualization.
Programming Languages: Python (Pandas, Numpy, Hugging Face, Tensorflow, Keras, NLTK, Scikit-Learn, Pytorch), R, Scala, NoSQL, Dax.
Data Engineering and DBs: Flask, Apache Spark, AWS Sagemaker, Pinecone, Faiss, Kafka, MongoDB, PostgreSQL, Elasticsearch, Logstash.
Generative AI Frameworks: Langchains, LLMs, AWS Bedrock, Claude, Stablediffusion, Llama, OpenAI.
Visualization Tools: Power BI, Power Apps, Tableau, Plotly, Kibana, Streamlit, MS Powerpoint, MS Excel.
Developer Tools: Git, Bitbucket, Anaconda, Jupyter Notebook, Google Colab, PyCharm.
Deployment Frameworks: Docker, Kubernetes, Airflow, AWS EC2, Azure Data Factory, Azure Databricks.

Yadav, K., Yadav, M., & Saini, S. (2021). Stock values predictions using deep learning based hybrid models. CAAI Transactions on Intelligence Technology, 7(1), 107–116. https://doi.org/10.1049/cit2.12052
Bilingual sentiment analysis for a code-mixed Punjabi English social media text. (2020, October 14). IEEE Conference Publication | IEEE Xplore. https://ieeexplore.ieee.org/document/9277309
Bi-LSTM and Ensemble based Bilingual Sentiment Analysis for a Code-mixed Hindi-English Social Media Text. (2020, December 10). IEEE Conference Publication | IEEE Xplore. https://ieeexplore.ieee.org/document/9342241
Siddiqui, A., Yadav, K., Dwivedi, S. (2016, July). Efficiency Enhancement Using Various Types Defected Ground Structures for Micro strip Patch Antenna. International Journal of Advanced Engineering Research and Applications. https://www.ijaera.org/manuscript/20160203003.pdf