Open to ML and MLOps roles

I build machine learning models and the pipelines that ship them.

I'm Ruchita, a B.Tech CSE graduate. I clean data, train and evaluate models with Python and Scikit-learn, then automate testing, security scanning, deployment and monitoring with Jenkins, GitHub Actions, Docker and AWS.

2ML projects with full CI/CD pipelines
0.878R² on the Mental Health Signal test set
3HIGH-severity CVEs found by Trivy and fixed
0.35MAE on a 0 to 10 score, so predictions land within about a third of a point

Projects

Both projects cover the full ML lifecycle, from data and modeling to deployment. A push to main runs the checks, builds the images and deploys to AWS automatically.

Oct 2026

Churn MLOps

End-to-end ML pipeline with CI/CD
  • A GitHub Actions pipeline lints, tests, retrains and validates the churn model, scans it and deploys it to AWS EC2 on every push.
  • A model quality gate (minimum ROC AUC plus regression checks) blocks weaker models, and MLflow tracks every experiment.
  • Trivy found three HIGH-severity vulnerabilities in a dependency; upgrading FastAPI fixed them.
  • Prometheus metrics from the prediction API feed Grafana dashboards for API performance and prediction drift.
PythonFastAPIMLflowDockerGitHub ActionsAWS EC2PrometheusGrafana

What runs on every push

  • Lint
  • Test
  • Retrain
  • Quality gate blocks weaker models
  • Security scan Trivy
  • Deploy AWS EC2
  • Monitor Prometheus + Grafana
Aug 2025 - Oct 2025

Mental Health Signal

End-to-end MLOps project
  • Explored a 5,000-record student dataset and trained a Random Forest regressor that predicts a mental health score (0 to 10) from 12 daily-habit features: R² 0.878, MAE 0.35 on held-out data.
  • Built one Scikit-learn Pipeline (log transform and scaling for numbers, ordinal and one-hot encoding for categories) so training and serving use identical preprocessing.
  • Compared three models and noted the Random Forest overfits (train R² 0.98), which tuning reduced but did not improve on test data.
  • Served with FastAPI behind Nginx and packaged as Docker images with Docker Compose.
  • A Jenkins pipeline runs on every GitHub push and deploys to AWS EC2, then runs a smoke test against the live server.
  • Security scan findings were fixed, and containers run as non-root with Docker health checks.
PythonScikit-learnFastAPIDockerJenkinsTrivyAWS EC2Nginx

Model comparison (test R²)

Linear Regression0.740
Random Forest, tuned0.865
Random Forest, deployed0.878

What runs on every push

  • Lint ruff
  • Test pytest
  • Build images docker compose
  • Security scan Trivy fails on HIGH / CRITICAL
  • Push Docker Hub
  • Deploy AWS EC2 over SSH
  • Smoke test /health
Mar 2026 - Apr 2026

Customer Behavior Analysis

End-to-end analytics workflow that turns raw customer data into insights on purchasing behavior, segments and loyalty, with SQL analysis and a Power BI dashboard.

PythonPandasPostgreSQLPower BI

Skills

The tools I have used in my projects, plus the ones I'm learning next.

Machine learning

PythonScikit-learnRegressionClassificationFeature engineeringModel evaluationRandom Forest

Data analysis

PandasNumPyEDASQLPostgreSQLPower BIJupyter

MLOps and DevOps

JenkinsGitHub ActionsDockerMLflowTrivyGitGitHub

Cloud and backend

AWS EC2LinuxFastAPINginxpytest

Monitoring

PrometheusGrafana

Learning next

KubernetesTerraform

Orchestrating containers and defining cloud infrastructure as code.

About me

I enjoy the whole life of a machine learning model: exploring the data, engineering features, training and evaluating models, and then making them testable, secure, deployable and observable.

I started in data and ML. During my Machine Learning internship I cleaned data and built regression and classification models with Scikit-learn. I then added the delivery side, building pipelines that check, containerize and deploy models without manual steps.

I'm now looking for a machine learning or MLOps engineer role where I can build models and keep them running reliably.

  • educationB.Tech CSE, JUIT
  • years2022 - 2026
  • cgpa7.7 / 10
  • locationIndia
  • statusOpen to work

Experience and education

Machine Learning Intern, Yhills Edutech Pvt. Ltd.

Jun 2025 - Jul 2025 | Remote
  • Built data cleaning and preprocessing workflows in Python and Pandas.
  • Trained and evaluated regression and classification models with Scikit-learn.

B.Tech, Computer Science and Engineering

Aug 2022 - Jun 2026 | Jaypee University of Information Technology | CGPA 7.7

Coursework: Machine Learning, Operating Systems, CN, DBMS, DSA, OOPS.

Certifications

Google Analytics (Skillshop), Sep 2026

Leadership at JUIT

Coordinator, Disciplinary Committee (Aug 2024 - May 2025). Design Manager, TIEDC Cell (Oct 2024 - Dec 2024).

Let's build something that works

I'm open to machine learning and MLOps roles, internships and full-time. Send me a message and I'll reply soon.