Machine Learning And Deep Learning Course

Using Generative Artificial Intelligence (AI) in Supply Chains

Supply Chain Foundations

Power Bi

Data science Mentorship Program

Tableau for Data Scientists

Deloitte Australia - Data Analytics Job Simulation

Bachelor of Engineering - BE

Master's degree

diploma in mechanical engineering

● Built a scalable time series forecasting pipeline that generated 5-year forecasts across 645 patent categories (IPC codes) using 31 years of historical data (1980–2010), enabling category-level trend prediction at scale. 
● Benchmarked 7 forecasting models across statistical (ARIMA, SARIMA, ETS) and machine learning (Linear Regression, Random Forest, XGBoost) and identified the best-performing model class per series by evaluating RMSE on a 5-year holdout (2011–2015), reducing model-selection effort across hundreds of series. 
● Engineered an end-to-end forecasting workflow including ADF stationarity testing on all 645 series, seasonal decomposition, hyperparameter tuning via grid search and AIC, and 3-lag feature engineering for ML models improving forecast reliability and producing diagnostic-driven model recommendations.

Research Intern (Data Science)

● Tracked supplier quality data across 100+ suppliers monitoring defect rates, rejection trends and corrective actions across procurement and production workflows and reporting findings to business partners through structured reporting models. 
● Managed a 4 stage inspection pipeline (raw material, post-machining, flow testing, final paint) monitoring cycle times, rejection patterns and quality deviations at each stage to flag production issues before they reached the client. 
● Maintained component-level traceability using heat number IDs across all production orders  identifying data quality issues such as missing heat IDs, incomplete test certificates and incorrect test readings and flagging discrepancies before client dispatch.

Quality Control Engineer

● Analysed daily transactional sales data across 5–6 months of records using SQL joining multiple tables, cleaning 
inconsistencies and identifying demand patterns and seasonal trends to support operational planning. 
● Built forecast vs actual reconciliation models comparing theoretical and actual consumption, helping improve alignment between expected demand and prep quantities and reducing over ordering on high-variance items. 
● Tracked waste trends and flagged high variance SKUs contributing to cost overruns, providing item level insight that supported weekly inventory and ordering decisions.

Operations Associate

Data Science Trainee / Intern

Data Analyst | Applied Data Scientist | Python, SQL, Power BI |  Supply Chain Analytics, Operations Analytics

github

Know about AKSHAY KANAVAJE's work experience, education, bootcamps, certifications, and much more. AKSHAY KANAVAJE