Navpreet Singh

Machine Learning Engineer

College Park, Maryland, United States6 yrs experience

Most Likely To SwitchHighly Stable

Key Highlights

Enhanced ML model reducing false positives by 11%.
Developed intelligent search system achieving 50% higher accuracy.
Optimized GRC application reducing query time by 40%.

Stackforce AI infers this person is a Machine Learning Engineer with expertise in Healthcare and Enterprise Software.

Contact

Skills

Core Skills

Machine LearningData EngineeringArtificial Intelligence (ai)Software DevelopmentDatabase Management

Other Skills

CommunicationComputer ScienceMachine Learning AlgorithmsData AnalyticsData MiningNatural Language Processing (NLP)Deep LearningComputer VisionPythonBack-End Web DevelopmentNode.jsSQLCUDABig DataObject Detection

Experience

Cigna

Machine Learning Engineer

May 2022 – Present · 3 yrs 10 mos · Boston, Massachusetts, United States · Remote

● Part of a 2-data-scientist team that enhanced the anti-fraud component of ML model.
● Reduced false positive rate by 11% and increased AUC by 2% by generating 2 new features to a dataset of 16M+ healthcare
insurance claims (baseline: 10 features).
● Implemented and deployed the ML model pipelines on Azure Machine Learning, utilizing Kubernetes service and Databricks storage to streamline the deployment process.
● Improved an urgent case detection system by 15% in both precision and recall through representation of OCR scanned documents with GPT-based language modeling and classification.

CommunicationComputer ScienceMachine LearningData Engineering

New york university

Research Assistant

Apr 2021 – Aug 2022 · 1 yr 4 mos · New York, United States

CommunicationComputer Science

Wissen technology

Machine Learning Engineer

Nov 2020 – Feb 2021 · 3 mos · India

● Worked on developing an intelligent search system for technical documents in the oil industry.
● Created datasets by processing large amounts of semi-structured and unstructured OCR scanned document text in Python.
● Fine-Tuned BERT transformer on local Q-A dataset, achieving 50% higher accuracy compared to traditional TF-IDF approach.

CommunicationComputer ScienceMachine LearningData Engineering

Metricstream

Software Engineer

Sep 2018 – Aug 2020 · 1 yr 11 mos · Bengaluru, Karnataka, India

● Developed an application for generating automatic reports for governance, risk and compliance management (GRC).
● Designed and developed data receiving module for GRC platform using Node.js and SQL for efficient report population.
● Optimized query performance, reducing time and memory usage by 40% and 80% respectively, resulting in improved user
satisfaction among 100+ users.

Back-End Web DevelopmentCommunicationSoftware DevelopmentDatabase Management