Summary
Overview
Work History
Education
Skills
Publications
Timeline
background-images
Nityam Churamani

Nityam Churamani

Bengaluru,Karnataka

Summary

AI/ML Analyst at JP Morgan Chase & Co. with 2+ years of experience in designing intelligent systems for crisis-time decision-making and firmwide resiliency across 300K+ critical assets. Skilled in agentic AI frameworks, NLP, LLMs, and quantitative modeling for real-world applications in finance and risk management. Achievements include constructing multi-agent AI systems for recovery planning automation, creating BERT-based models for workforce resiliency mapping with over 96% accuracy, and spearheading AI-driven trade surveillance projects to improve fraud detection in Fixed Income, FX, and Commodities assets by reducing false positives by 25%. Published research on cancer therapy prediction and financial forecasting using sentiment analysis. Aspiring to pioneer scalable, explainable AI systems in high-stakes sectors like finance and technology.

Overview

2
2
years of professional experience

Work History

Applied AI/ML Analyst

JP Morgan Chase & Co.
09.2024 - Current
  • Project: Agentic Resiliency Management
  • Spearheading an AI-driven agentic system to simulate, reason, and validate over 3,000 recovery strategies for critical assets (staff, sites, applications), ensuring operational continuity during disruptive events.
  • Utilizing state-of-the-art agentic AI frameworks like AutoGen and LangChain to build multi-agent systems with autonomous planning, dialogue-based reasoning, and strategy evaluation.
  • Architected intelligent agents that emulate crisis-time decision-making, reducing manual risk analysis across 5 lines of business and strengthening resiliency for 300K+ critical assets—projected to mitigate substantial financial losses through enhanced firmwide disaster recovery preparedness.
  • Project: Mapping Unmapped Staff to Business Resiliency Plans
  • Explored and evaluated machine learning and deep learning models, including autoencoders, artificial neural networks, and XGBoost, to address the staff-to-resiliency-plan mapping problem.
  • Developed a BERT-embedding-based similarity search method providing top-3 recommendations, achieving 96%+ mapping accuracy in validation.
  • Automated resiliency mapping workflows, reducing manual labour and improving regulatory compliance.

Applied AI/ML Analyst

JP Morgan Chase & Co.
06.2023 - 08.2024
  • Project: AI-Driven Trade Surveillance
  • Built supervised ML pipelines (XGBoost, Logistic Regression) to detect spoofing patterns across 10M+ transactions, enhancing fraud detection across Fixed Income, Swaps, Commodities, and FX.
  • Conducted validation of 300+ financial risk features for mathematical and logical integrity, enhancing reliability in predictive models.
  • Designed and deployed the Investigator Score, a quantitative risk metric, increasing alert capture precision by 30% and aiding decisions for 100+ compliance officers.
  • Applied NLP techniques (TF-IDF) on 4,000+ investigator comments to uncover 5 manipulation indicators(Time, Volatility, Volume, Price and Momentum), reducing manual alert verification by 30%.

Software Engineer Intern (Data Science)

JP Morgan Chase & Co.
02.2023 - 05.2023
  • Project: AI-Driven Trade Surveillance
  • Selected through a pan-India hackathon as 1 of 600 interns to join the Applied AI/ML Trade Surveillance team managing 400M+ Dollars in Fixed Income assets.
  • Collaborated with 5 cross-functional business teams to enhance financial risk identification processes by 20%.
  • Developed and validated 50+ financial risk features, improving model performance and predictive accuracy by 25%.

Education

B.Tech - Computer Science

PES University
Bengaluru, India
08.2023

Class XII -

Narayana E-Techno School
Bengaluru
03.2019

Skills

  • Large Language Model
  • Agentic Artificial Intelligence
  • Python
  • Numpy
  • Pandas
  • Pytorch
  • Statistics
  • Machine Learning
  • Data Science
  • BERT
  • PySpark
  • Transformers
  • Deep Learning
  • Natural Language Processing

Publications

  • Computational Techniques for Predicting Response to Therapy for Cancer, May 2022 – May 2023

Developed predictive models using supervised and unsu-

pervised ML techniques—including XGBoost, Elastic Net,

and Neural Networks—combined with dimensionality re-

duction to analyze cancer gene expression data and im-

prove personalized drug response (IC50) predictions. Research Publication Link- https://ieeexplore.ieee.org/document/10392026


  • Stock Price Prediction Using Machine Learning and Sentiment Analysis, May 2021 – June 2022

Utilized LSTM and ARIMA models alongside NLP-based

sentiment analysis (e.g., VADER) to predict stock price

movements, integrating deep learning and natural language processing for enhanced financial forecasting. Research Publication Link- https://ieeexplore.ieee.org/document/10094422

Timeline

Applied AI/ML Analyst

JP Morgan Chase & Co.
09.2024 - Current

Applied AI/ML Analyst

JP Morgan Chase & Co.
06.2023 - 08.2024

Software Engineer Intern (Data Science)

JP Morgan Chase & Co.
02.2023 - 05.2023

B.Tech - Computer Science

PES University

Class XII -

Narayana E-Techno School
Nityam Churamani