Research Assistant — California Transportation Analytics Program (CalTAP)
Applied Research | Transportation Analytics | Data-Driven Policy Systems
Conducted applied data analytics and AI-assisted research to support intelligent transportation systems, translating complex technical findings into decision-ready insights for policy makers, planners, and non-technical stakeholders.
- Synthesized peer-reviewed research papers, evaluation reports, and large-scale real-world transportation datasets to identify patterns, risks, and optimization opportunities in urban mobility and infrastructure planning.
- Authored data-driven research briefs, analytical articles, and stakeholder newsletters, prioritizing model explainability, interpretability, and transparency to ensure actionable and trustworthy insights.
- Supported iterative research refinement by aligning analytical outputs with evolving performance metrics, regulatory frameworks, and public policy objectives, ensuring relevance across changing requirements.
- Applied statistical analysis, data visualization, and exploratory data analysis (EDA) techniques to clearly communicate trends, trade-offs, and system impacts.
- Collaborated with interdisciplinary teams to bridge data science, policy, and real-world implementation, strengthening the end-to-end research-to-decision pipeline.
Core Technologies & Skills
Data Analysis, Applied Machine Learning, Explainable AI (XAI), Research Analytics, Transportation Systems, Urban Analytics, Statistical Modeling, Data Visualization, Policy Analytics, Technical Writing, Stakeholder Communication, Decision Intelligence
AI/ML Intern — Finlogic Technologies India Pvt. Ltd.
Machine Learning | Computer Vision | Predictive Analytics | AI Automation
Designed and deployed LLM-based intelligent agents for document understanding and task automation, improving operational efficiency and reducing manual intervention.
- Built predictive models using XGBoost and Monte Carlo simulations, achieving 97% accuracy in financial forecasting and decision-support systems.
- Developed computer vision pipelines leveraging ConvNeXt and OCR technologies, reducing human effort by 70% through automated document processing.
- Engineered end-to-end ML systems, encompassing data preprocessing, feature engineering, model training, API integration, and performance tuning for production deployment.
- Optimized inference pipelines, reducing latency by 30%, enabling real-time insights for financial applications.
- Implemented robust document classification systems, including a "does not belong" routing mechanism, enhancing model reliability and adaptability in dynamic data environments.
- Created automated check processing and signature verification pipelines, applying image processing, Hough transforms, edge detection, and anchor-based rules to ensure high data accuracy and fraud prevention.
- Integrated real-time financial APIs for dynamic stock and mutual fund recommendations, incorporating portfolio optimization under multiple risk profiles (aggressive, moderate, conservative).
Core Technologies & Skills
Machine Learning, Deep Learning, LLMs, Computer Vision, OCR, Predictive Modeling, Data Preprocessing, Feature Engineering, API Integration, XGBoost, Monte Carlo Simulation, ConvNeXt, Portfolio Optimization, Real-Time Data Systems, Automation Pipelines
AI Developer & Generative AI Intern — Daira EdTech Pvt. Ltd.
Generative AI | NLP | Speech-to-Text | Text-to-Speech | Machine Learning | Data Science
Developed Generative AI Roadmap Web Platform, generating personalized learning and career roadmaps for students across 200+ fields. Optimized prompt engineering strategies for actionable outputs tailored to Master's, startup, or research paths.
- Built Speech-to-Text Pronunciation Assessment Tool using Whisper API and alternative free STT APIs, enabling students to improve pronunciation accuracy by ~35% and track multi-stage progress.
- Designed Text-to-Speech Learning Tool for Children, integrating speech recognition APIs and instant feedback mechanisms, achieving 80%+ correct word recognition after three practice sessions.
- Refactored ML pipelines for 78% faster training and scalable deployment, improving system performance for user engagement analytics.
- Translated insights from 5,000+ users into actionable recommendations, connecting AI outputs with business decisions.
- Addressed API limitations, optimized scoring algorithms, and handled diverse user inputs, ensuring high reliability and generalization across educational scenarios.
- Delivered 500+ personalized career and educational roadmaps, supporting data-driven decision-making for students and increasing platform adoption.
Core Technologies & Skills
Generative AI, NLP, Prompt Engineering, Speech-to-Text, Text-to-Speech, Data Science, Machine Learning, Pipeline Optimization, Python, API Integration, User Behavior Analytics, Educational AI Tools, Scalable ML Systems
Get in Touch
Open to full-time roles and internships in AI, Machine Learning, and Software Engineering. Recruiters and hiring managers can reach me directly at niyatikapadia111@gmail.com or via LinkedIn
Experience
Showcasing comprehensive Machine Learning, Data Science, and AI portfolio implementations featuring predictive modeling, computer vision, NLP, and scalable software development solutions.