Lekha Challappa Email & Phone Number

Louisville, Kentucky, United States

Render Capital Venture PartnerNVIDIA Inception Startup.At Aurat, we’re not just building aroma; we're engineering a transformation in your workday. Our scientifically designed aromatics do more than fill a room with fragrance—they're meticulously formulated to modulate neurotransmitters and elevate productivity. Join us on a journey that transcends time, where ancient art meets modern science, and discover how Aurat is redefining the essence of productivity.

Atlanta, Georgia, United States

Domain Anchorage in GPT-4: A Computational Linguistic Analysis of Lexicographic Profiling and Its Implications for Unintended Information DisseminationAuthors: Lekha Challappa, Jenevieve Zhang, and Rajiv Garg PhDAbstract: Our study expands upon recent work explaining in-context learning as implicit Bayesian inference, where language models infer shared latent concepts from examples. We analyze GPT-4’s semantic attention post-domain priming, using computational linguistics to quantify response similarity to lexicographically independent queries with the same intent. We assess the potential for privacy breaches due to domain anchorage, examining how attention and embedding layers process linguistic patterns. We hypothesize that domain-specific words receiving higher gradient updates can introduce bias, create semantic echo chambers, and oversimplify relationships. Grounded in Mohamed Zakaria Kurdi’s frameworks, this research uses lexical, semantic, syntactic, and positional similarities to analyze GPT-4’s vector transformations and attention distributions. By simulating domain-specific interactions through declarative primes and interrogative inputs, we highlight significant privacy and ethical concerns, as the model may share information across users due to domain anchorage.Recognition: Abstract presented at POMS (Production & Operations Management Conference) 2024Publication: HICSS 2025

Atlanta, Georgia, United States

Integrated cognitive computing technologies within NCR VOYIX to revolutionize customer experiences in retail, restaurants, and digital banking. My current project harnesses the power of Large Language Models (LLMs), via Microsoft Azure and Databricks, to augment the NCR knowledge repository and elevate customer support paradigms.Key Project Initiatives: -AI Collaboration: Fostering a synergistic loop between AI analytics and human expertise to refine communication and decision-making capabilities.-Efficient AI Solutions: Focused on optimizing resources for maximum AI performance without compromising on cost-efficiency.-Quality Assurance in AI: Establishing stringent measures to monitor and correct AI predictions, maintaining the integrity of customer interactions.

Atlanta, Georgia, United States

Master's in Management Program: Applicant Sourcing Research Ass.- Marketing AI Tool Design w/ Prof. Usha Rackliffe: 1. We will initiate the recruitment optimization by employing the CRISP-DM methodology to aggregate and preprocess multifaceted applicant data using PySpark, ensuring a robust feature set through techniques like categorical encoding and normalization. 2. Our model selection will prioritize Logistic Regression and Random Forest classifiers within PySpark's MLlib to facilitate a baseline for classification accuracy and interpretability, further refined by a systematic hyperparameter tuning process using cross-validation techniques.3. Post model validation, we will deploy the most performant algorithm into a production environment, integrating it with marketing APIs to dynamically score and segment potential applicants, thus enhancing the precision of our targeted enrollment strategies.4. Continuous monitoring of model efficacy will be implemented through performance metrics such as AUC-ROC and F1 Score, ensuring that the predictive model adapts to evolving applicant profiles and sustains its accuracy in identifying prospective students likely to enroll.

Elizabethtown, Kentucky, United States

Utilized insights from the article "ACCF/AHA Guidelines for the Management of Patients with ST-Elevation Myocardial Infarction" to build a sophisticated riskprediction ML model for interventional cardiology post COVID-19 with IV Cardiology research team.• Employed advanced tools and technologies including Python, scikit-learn, and TensorFlow for machine learning, along with R for statistical analysis.• Liaised with physicians to understand clinical needs and translated quantitative insights into actionable information, improving patient management andoutcomes.• Analyzed Epic patient data to predict risk factors and outcomes, resulting in a 35% improvement in workflow efficiency through enhanced predictive analysis..A. Primary IDE: MySQL & Microsoft SQL ServerB. worked to organize electronic health records C. learned SQL techniques to import ECG to data to HIMS systems database 4. used knowledge of SQL querying from data analytics courses to assist practice research manager

Master Of Science - Ms, Business Analytics: Ai Track

Activities and Societies: Student Leadership Council Role: MSBA Career LiasonGoizueta Business School's STEM-credentialed Master.

High School Diploma

Bachelor Of Science - Bs, Health Science -Health Administration Minor: Health Management And Policy And Public Health

Activities and Societies: Foote Fellow Honors Society Presidential Scholarship Award Recipient uStart Entrepreneurship-Pitch Competition.