Ayushi Mandlik

Data Scientist | ML Engineer

Profile

Resourceful and analytical Data Scientist with a strong foundation in machine learning, statistical modeling, and data engineering. Experienced in delivering real-world solutions across domains such as logistics, customer analytics, and scientific research. Skilled in building and deploying predictive models, performing advanced data analysis, and automating data pipelines. Proven track record of cross-functional collaboration, stakeholder engagement, and leading technical discussions. Demonstrated leadership through technical coordination roles and hackathon success.

Technical skills

Languages & Libraries:

Python (NumPy, Pandas, Scikit-learn, TensorFlow, Keras), SQL, Bash, C++

ML & Statistical Methods:

Supervised/Unsupervised Learning, Regression, Classification, Clustering (DBSCAN, K-Means), Forecasting, CNNs, Bayesian Inference, A/B Testing

Data Tools & Platforms:

Dataiku DSS, GCP (BigQuery), AWS (Foundational), Git, Tableau, Zoho CRM

Engineering & MLOps:

ETL, Workflow Automation, HPC (Slurm, OzStar, JURECA), Jupyter Notebooks

Visualization & Reporting:

Tableau, Matplotlib, Seaborn, Dashboards, Technical Presentations

Relevant experience

CRM Workflow Automation & Customer Insights

Data Analyst / Project Lead – Truetel, Melbourne | Apr 2024 – Mar 2025

Software & Tools: Python, SQL, Excel, Zoho CRM, Google Sheets

→ Developed automated lead management and customer engagement workflows using Zoho CRM.
→ Implemented data pipelines to migrate and clean customer data, improving CRM usability and campaign tracking.
→ Applied rule-based customer segmentation for improved targeting and engagement.

Strategic Relevance: Showcases customer segmentation, ETL workflows, and cross-team collaboration.

Forecast Optimization for Parcel Logistics

Data Science Intern – Australia Post, Melbourne | Apr 2024 – Dec 2024

Software & Tools: Python, Dataiku DSS, GCP (BigQuery), Tableau, Excel, SQL, Git

→ Enhanced parcel volume forecasting using hierarchical reconciliation; improved accuracy (MAPE reduction of 1%).
→ Built automated pipelines for demand forecasting and reporting within Dataiku DSS.
→ Conducted A/B comparisons for model evaluation and communicated results via Tableau dashboards.

Strategic Relevance: Highlights statistical analysis, A/B testing, forecasting, and data visualization.

Real-Time CNN Pipeline for Radio Astronomy

Research Project – Centre for Astrophysics and Supercomputing, Melbourne | Oct 2019 – Jan 2024

Software & Tools: Python, TensorFlow, Keras, NumPy, SciPy, Slurm, HPC

→ Developed a GPU-accelerated CNN pipeline for real-time classification of astrophysical signals.
→ Applied DBSCAN clustering to reduce RFI noise and streamline input to classifiers.
→ Conducted regression modeling to improve detection rates across varying input conditions.

Strategic Relevance: Demonstrates model building, MLOps practices, clustering, and noise filtering.

Statistical Modeling & Bayesian Inference for Pulse Fitting

Research Project – Centre for Astrophysics and Supercomputing, Melbourne | Oct 2019 – Jan 2024

Software & Tools: Python, NumPy, SciPy, Pandas, Matplotlib, Jupyter

→ Used Bayesian inference for parameter estimation and model comparison.
→ Evaluated models using AIC, chi-square, and likelihood-based criteria.

Strategic Relevance: Demonstrates understanding of statistical modeling and risk-based evaluation.

Large-scale Signal Processing for Galaxy Imaging

Masters Researcher – Max Planck Institute for Radio Astronomy, Bonn, Germany | Oct 2017 – Jan 2018

Software & Tools: Python, JURECA HPC, NumPy, Astropy, Matplotlib, Git

→ Processed 100+ TB of astronomical data using Python pipelines for galaxy imaging and magnetic field analysis.
→ Built 3D visualizations and automated scripts to analyze star-forming regions and distance estimation.

Strategic Relevance: Strong ETL, large-scale data processing, and scientific computing experience.

Data Visualization of Star-Forming Regions

Research assistant – Max Planck Institute for Radio Astronomy, Bonn, Germany | Mar 2017 – Oct 2017

Software & Tools: Python, NumPy, Astropy, Matplotlib

→ Built 3D visualizations to analyze galactic star-forming regions and identify distance-related patterns.
→ Integrated multi-source datasets to support spatial analysis and improve interpretability of key variables.

Strategic Relevance: Statistical Analysis, Strong communication and presentation skills

Awards and scholarships

Centre for Supercomputing and Astrophysics

Scholarship for thesis project

Max Planck Institute for Radio astronomy

Scholarship for Master thesis project

Christ University

Scholarship for being in top three students

Ayushi Mandlik — ayushimandlik09@gmail.com — (+61) 481015577