Ashutosh Rane Email & Phone Number

United States

United States

• Worked with one of the biggest insurance brokerage company in the world.
• Developed and implemented a Claims Prediction Model using Scikit-learn and PySpark, streamlining claim triaging and automating approval processes, which reduced claim processing time by 20% and achieved close to 80%.
• Built a Customer Churn Model leveraging advanced predictive analytics to analyze client behaviors, identify at-risk customers, and provide actionable insights for enhancing client retention strategies, contributing to.
• Conducted time series analysis on historical claims data, predicting seasonal trends and identifying anomalies, leading to better-informed risk mitigation strategies.
• Streamlined data pipelines for claims and policy data using PySpark and Databricks, ensuring scalability, reliability, and improved data ingestion for analytics workflows.

United States

• Fine-tuned GPT-3.5 Turbo and GPT-4 on a custom dataset to enforce rule-based behavior, ensuring consistent and high-quality outputs aligned with specific requirements
• Developed an in-house Text-to-Speech (TTS) model using CoquiTTS, integrated it into a client-ready API, enhancing accessibility and delivery capabilities
• Designed and implemented benchmarks for testing outputs of various Large Language Models (LLMs), and created a novel LLM-based evaluation system to serve as a judge for comparing and ranking output quality

• Developed OCR models achieving 96% accuracy to extract essential data from sensitive documents, built secure REST APIs for government document verification, and deployed OCR API services on AWS EC2 servers.
• Orchestrated deployment and management of machine learning models on AWS using Docker and Docker Swarm, ensuring robust data security, authentication, and scalability while maintaining 99.9% server uptime.
• Implemented ML models for liveness detection, facial recognition, anti-spoofing, selfie segmentation, captcha breaking, and document/card detection, significantly enhancing verification systems for banking and other.
• Spearheaded the development of a self-hosted cloud token-based central authentication system for the ML API suite, ensuring secure and streamlined access management across multiple services.
• Worked in an Agile environment with tools such as GitHub, JIRA maintaining CI/CD pipelines and collaborating effectively within cross-functional teams.

Master Of Science - Ms, Electrical And Computer Engineering

Bachelor Of Technology - Btech, Electrical, Electronics And Communications Engineering

Good Knowledge of Python,R ,Machine learning,Neural Networks.

Bachelor Of Engineering - Be

Hsc

Ssc