Open to opportunities

Khushi User

@khushiuser10

Message

Computer Science undergrad building end-to-end AI/ML and GenAI products with Python, Streamlit, and Gemini APIs.

India

Message

What I'm looking for

I’m looking for an entry-level role where I can ship AI/ML features end-to-end—data pipelines, model monitoring, and explainability—using Python and GenAI, and learn in a collaborative team that values measurable impact.

I’m a Computer Science undergraduate who builds end-to-end AI/ML and Generative AI applications, turning messy data into measurable outcomes. I focus on practical engineering—clean pipelines, strong feature engineering, and reliable integrations—so models work in real products, not just notebooks.

In my Customer Churn Predictor, I engineered 4 domain-driven features and used XGBoost with class-imbalance handling to achieve ROC-AUC 0.846 and churn recall 0.81. I also used SHAP explainability (waterfall + summary plots) to pinpoint key churn drivers and translate them into actionable business recommendations.

I care about model health after deployment. I built a monitoring dashboard that performs KS-Test drift detection against the training baseline, logs predictions to SQLite, and visualizes churn risk trends—then deployed it on Streamlit Cloud for an easy workflow.

Beyond classical ML, I create GenAI products like HireBot, a resume screener that uses the Gemini API via LangChain to produce structured JSON outputs (match score, ATS score, and skill gaps). I also developed AI-powered roadmap and interview-prep modules, plus JobLens, a data & AI analytics dashboard driven by an ETL pipeline fetching 160+ live listings and powering interactive Plotly visualizations.

Experience

Work history, roles, and key accomplishments

Customer Churn Predictor

Built an end-to-end churn ML pipeline on 7,043 telecom records with class-imbalance handling and tuned XGBoost to ROC-AUC 0.846 and churn recall 0.81. Engineered domain-driven features, applied SHAP explainability to identify key churn drivers, and deployed a Streamlit monitoring dashboard with KS-test drift detection.

Python XGBoost Streamlit SHAP Explainability Data Pipelines Feature Engineering Model Monitoring SQLite REST APIs

AI Resume Screener

HireBot

Developed a GenAI resume screener that analyzes resumes against job descriptions using the Google Gemini API and returns structured JSON with match score, ATS score, and skill gaps. Built PDF parsing with PyMuPDF, prompt orchestration with LangChain, and visual skills matching with Plotly, plus an interview question generator with AI answers and score history in SQLite.

Python Streamlit Gemini API PyMuPDF SQLite Plotly REST APIs Prompt Engineering NLP Lang Chain

Job Market Analytics

JobLens India

Created a real-time JobLens dashboard tracking Data & AI hiring trends across India using an ETL pipeline that fetches 160+ listings from JSearch via RapidAPI. Implemented pagination/rate-limiting, data cleaning and deduplication in Pandas, structured the dataset in SQLite, and deployed an interactive Streamlit app with Plotly city-wise hiring and company rankings.

Python Pandas SQLite Streamlit Plotly ETL Data Cleaning API Integration Data Visualization GitHub

Education

Degrees, certifications, and relevant coursework

Galgotias College of Engineering and Technology

B.Tech, Computer Science and Design

2023 - 2027

Grade: CGPA 7.4

Activities and societies: HackFest participation (GeeksforGeeks); certifications in SQL (Basic) via HackerRank; GitHub Foundations (GitHub Skills).

Pursuing a B.Tech in Computer Science and Design (2023–2027) with CGPA 7.4. Coursework includes DSA, DBMS, Operating Systems, Computer Networks, OOP, Machine Learning, and Software Engineering.