Soniya Phaltane - AI & ML Engineer

About Me

I am a Computer Science graduate student at Santa Clara University specializing in Artificial Intelligence and Machine Learning. With hands-on experience in developing RAG models and working with AWS, I've developed a strong foundation in modern AI technologies.

My professional experience includes working as a Machine Learning Intern at TechR Business Solutions, where I implemented an LSTM-based anomaly detection model to analyze system logs in real-time.

I'm currently seeking Summer 2025 internship opportunities as a Data Science or Machine Learning Intern, where I can apply my skills and continue to grow in the field of AI.

ML Frameworks & Tools

PyTorch Keras scikit-learn Hugging Face RAG AWS SageMaker

ML Algorithms & Models

DBSCAN Random Forest CNN RNN Decision Tree Regression SVM

Programming & Technologies

Python Docker Kubernetes AWS Java React Node.js

Work Experience

Machine Learning Intern

TechR Business Solutions, Pune, India

February 2023 - April 2023

Implemented an LSTM-based anomaly detection model to analyze system logs in real-time, automatically identifying unusual patterns with 85% accuracy.
Built data preprocessing pipeline in Python to transform raw log data into sequential features for the LSTM model.
Integrated Owasp Zap scanner with the deployment pipeline to automate security testing.

Education

Master of Science in Computer Science

Santa Clara University, Santa Clara, CA

Sept. 2024 – June 2026

Bachelor of Engineering in Computer Science

Modern College of Engineering, Pune University, Pune, India

Sept. 2020 – May 2024

Projects

🤖 AI & Machine Learning

K8shield - Kubernetes Security Monitoring

A security monitoring system for Kubernetes clusters with ML-based anomaly detection. Integrates real-time metrics, compliance checks, and automated threat detection.

Kubernetes Machine Learning AWS EC2 Security Python

View Project

RAG-Based YouTube Study Assistant

AI assistant that analyzes YouTube educational videos to automatically generate detailed summaries, flashcards, MCQs, and learning resources using RAG with Gemini API, FAISS, and sentence transformers.

RAG Gemini API FAISS Python NLP Vector Search

View Project

Aisnoop - AI Tools Web Scraper

Web scraper tool that automatically finds and categorizes free AI tools from articles across the web, helping users discover new AI resources efficiently.

Python Web Scraping NLP Automation

View Project

Sentence-Transformer Image Matcher

A simple image description matching system using SentenceTransformers to find the most similar image based on textual descriptions through semantic similarity.

Python SentenceTransformers NLP Computer Vision

View Project

Realtime Image Generation via Audio

Developed an end-to-end AI pipeline orchestrating multiple OpenAI APIs for audio-to-image generation and cached image features to optimize API usage through image feature matching.

OpenAI APIs Whisper API ChatGPT API DALL-E API PostgreSQL

View Project

📊 Data Science & Analytics

Customer Return Intent Prediction

Developed a machine learning model to predict restaurant customer return probability by analyzing publicly available restaurant review datasets.

Random Forest Ensemble Learning Logistic Regression Decision Tree

View Project

💻 Systems & Networking

OS-prioritypipes - xv6 OS Enhancement

Experimented with the xv6 operating system by adding a custom timer and implementing RMS (Rate Monotonic Scheduling) and Priority scheduling algorithms.

C Operating Systems xv6 Scheduling Systems Programming

View Project

Custom UDP Protocol Implementation

Client-server applications with custom protocols built on UDP, featuring error handling, acknowledgment mechanisms, retry logic, and network access verification with subscriber databases.

C UDP Socket Programming Network Protocols Error Handling

View Project

Hackathons

Neighborhood Net (HopeSF)

Digital platform connecting homeless individuals in San Francisco with real-time resources including food, shelter, and hygiene services through an interactive map, provider dashboard, and AI chatbot.

React TypeScript Supabase Mapbox OpenAI Civic Tech

View on Devpost

ReportWeave Dashboard

Automated dashboard solution for generating comprehensive reports with data visualization and analytics capabilities, streamlining report creation workflows.

React Dashboard Data Visualization Analytics

Watch Demo

CodeShield - Privacy-Preserving AI Coding

Privacy-preserving AI coding assistant for enterprises that abstracts codebase structure for AI analysis while keeping proprietary code private. Integrates Semgrep security scanning and AWS Bedrock.

Python AWS Bedrock Claude API Semgrep Security React

View on Devpost

Academic Publications

Kubernetes Application Monitoring System - An Review

Journal of Engineering Computing and Architecture, ISSN No. 1934-7197

Read Publication

Customer Return Intent Prediction using Machine Learning And Data Science

Journal of Engineering Computing and Architecture, ISSN No. 1934-7197

Read Publication

AI Enhanced - Kubernetes Application Monitoring System

International Research Journal of Engineering and Technology, ISSN No. 2395-0056

Read Publication

Get In Touch

🔗

🐙

GitHub

📄

Resume