Manmohan Singh Email & Phone Number

Newport Beach, California, United States

Responsibilities:• Developed scalable machine learning models using Python and TensorFlow for predicting credit risk based on users' financial data, incorporating LLMs for feature extraction.• Developed abstractive summarization for documents with unlimited size using LLM - Llama2 and LangChain.• Utilized Generative AI techniques to create synthetic datasets for training, addressing data privacy concerns and improving model robustness against diverse financial scenarios.• Developed RESTful APIs using Flask to serve the ML models, enabling real-time credit risk assessments in Experian's digital platforms.• Developed a monitoring framework for model performance and bias detection using AWS CloudWatch and custom Python scripts for data analysis.• Set up model monitoring frameworks using Prometheus and Grafana for real-time monitoring of model performance and bias detection, ensuring models remain fair and accurate over time.• Implemented data encryption and secure data access mechanisms to protect sensitive financial information in compliance with GDPR and other regulatory standards.• Worked closely with the cybersecurity team to assess and mitigate potential security risks associated with deploying AI models.• Deployed HuggingFace models in AWS cloud using Terraform as part of MLOps.• Implemented AWS Step functions to automate and orchestrate the Amazon SageMaker related tasks such as publishing data into S3, training ML model and deploying it for prediction.

Hartford, Connecticut, United States

Responsibilities:• Collaborated with cross-functional teams to design data-driven solutions for healthcare and pharmaceutical customers.• Designed and prototyped ML based recommender (Keras) for optimized specialist referrals and faster diagnoses.• Worked on data mining tasks and data wrangling tasks on chat transcripts to convert and fed data into algorithms using python.• Achieved a validation R2 of 98% on predicting the ingredient cost of a drug by building Machine Learning models on claims data.• Built Machine learning classification models to identify the positive and negative drivers impacting the HCP Rx referrals based on the hcp demographics, prescription history, patient profile and the drug marketing sales data.• Utilized various machine learning models to implement the use cases and developed interactive dashboards. Additionally, employed natural language processing techniques to extract and analyze business data within APM.• Established CI/CD pipelines using Jenkins for automated testing, building, and deployment of ML models and their dependencies to Azure Kubernetes Service (AKS).• Data-preprocessing tasks included converting textual information into tabular data with the help of machine learning algorithm libraries like NLTK for NLP projects, OpenCV for computer vision projects and worked with other machine learning preprocessing techniques like dimensionality reduction, feature selection and other data cleansing using python libraries NumPy, Pandas, auto-encoders.• Performed data cleaning EDA using seaborn on the data to find useful insights, outliers, and missing values on structured and unstructured data.

Detroit, Michigan, United States

Responsibilities:• Collaborate with software engineers and data engineers to seamlessly integrate the prediction system with Financial Services' existing infrastructure and databases, version control systems Used Git/GitHub for collaborative development.• Develop monitoring mechanisms using Python libraries for tracking model performance and data distribution drift.• Set up alerting mechanisms using Python libraries or services like AWS CloudWatch to notify stakeholders of anomalies in prediction trends.• Document the entire data science pipeline using Markdown files, covering data preprocessing, model development, and deployment procedures.• Created the Linux services to run REST web services using shell script.• Creating AWS S3 buckets, performed folder management in each bucket, managing cloud trail logs and objects within each bucket.• Automating the existing scripts for performance calculations using NumPy and SQL alchemy.• Defining branching, labelling, and merge strategies for all applications in Git.• Configuring Elastic Load Balancers with EC2 Auto Scaling groups.• Worked on project driven on AWS Connect where we enable communication between Lex, Lambda, and customer.• Used Python (NumPy, SciPy, Pandas, Scikit-Learn) to develop variety of models and algorithms for occasionally real-time analytic purposes.• Knowledge using CI/CD development practices.• Worked on various machine learning algorithms and statistical modelling like decision trees, regression models, neural networks, SVM, clustering to identify Volume using Scikit-learn package in python, MATLAB.• Created Machine Learning Jobs from Kibana X-Pack ML component for anomalies and watcher for ML jobs for root-cause analyses to predict KPI Metrics and errors.• Worked on conversion of structured and non-structured data.

India

Responsibilities:• Participated in all phases of data mining, data collection, data cleaning, developing models, validation, and visualization to deliver data science solutions.• Gathering Requirements from FIU (Financial Intelligence Unit) and Regulators. Developed advanced analytical models and computational solutions using large-scale data manipulation and transformation, statistical analysis, machine learning, visualization.• Collaborated with data engineers and operation team to implement ETL process, wrote and optimized SQL queries to perform data extraction to fit the analytical requirements.• Explored and analyzed the customer specific features by using Matplotlib, Seaborn in Python and dashboards in Tableau.• Used RMSE/MSE to evaluate different models' performance. Worked closely with internal stakeholders such as business teams, product managers, engineering teams, and partner teams.• Designed database solution for applications, including all required database design components and artifacts.• Generate reports to meet regulatory requirements.• Worked on data cleaning and ensured data quality, consistency, integrity using Pandas, NumPy. Experience in Deep Learning frameworks like MX Net, Caffe 2, TensorFlow, Theano, CNTK, and Keras to help our customers build DL models. • Creating statistical models using distributed and standalone models to build various diagnostics, predictive and prescriptive solution.• Performed data imputation using Scikit-learn package in Python.• Used RMSE/MSE to evaluate different models' performance.• Create different charts such as Heat maps, Bar charts, Line charts etc.

India

Responsibilities:• Designed and developed the server module and resolved issues and responsible for its enhancements.• Architected and developed Python and Django for the backend development and front-end application using React, Webpack, Redux, and ES6/7 and PostgreSQL for the database.• Worked on Django ORM module for signing complex queries.• Load balancing of ETL processes, database performance tuning ETL processing tools.• Loaded the data from Teradata to HDFS using Teradata Hadoop connectors.• Defined different Django API profiling techniques for faster rendering information.• Used Ansible, Vagrant, and Docker for managing the application environments.• Developed Python scripts for testing the car functioning on different services.• Implemented various Validation Controls for form validation and implemented custom validation controls using Object Oriented JavaScript.• Developed Restful Microservices using Flask and Django and deployed on AWS servers using EBS and EC2.• Developed views and templates with Django view controller and template to create a user-friendly website interface.• Worked on java developing or java programming/scripting, Django and RabbitMQ queues.• Used Circle CI for continuous integration and delivery platform over GIT.• Used Django Migrations for designing and expanding of the database.• Automated most of the daily task using python scripting.• Involved in the CI/CD pipeline management for managing the weekly releases.• Worked on Jira for managing the tasks and improving the individual performance.• Making recommendations to the team in terms of appropriate testing techniques, shared testing tasks.