Mohit Paratane

Strategic Consultant known for high productivity and efficiency in task completion. Specialize in business process optimization, risk management, and stakeholder engagement. Excel in problem-solving, communication, and adaptability to ensure successful project outcomes. Navigate complex challenges using analytical skills and collaborative approach to drive organizational improvements.Proven ability to deliver actionable insights and improve response times, ensuring compliance with regulatory standards and driving data quality initiatives.

• Analysis of the Impact of COVID-19 on the UK Airline Industry, 04/01/23, 09/01/23, Python, Time Series Forecasting, Strategic & Financial Analysis, Analysed 1,600+ days of flight and COVID-19 data using Python to measure the pandemic’s impact on air travel; identified a 90% drop in flights during peak lockdowns and developed SARIMAX and LSTM models with over 85% accuracy to forecast industry recovery. Estimated £20B+ in financial losses using time-series and trend analysis, and applied PESTEL, SWOT, and Porter’s Five Forces to deliver a strategic report outlining operational risks and long-term recovery recommendations.
• Used Car Market Analysis – BMW 1 Series, 01/01/23, 03/01/23, Excel, SPSS, Regression Modelling, Data Visualisation, Collected and analysed 349 used car listings to identify key pricing factors; built a regression model (R² = 0.816) that reduced price estimation errors by 18%, validated results through hypothesis testing, residual analysis, and simulated A/B testing scenarios to support pricing decisions. Produced a pricing tool and insight report highlighting ideal value ranges (£15K–£21K) by fuel type, enabling more accurate pricing decisions for dealers and buyers.
• Speech & Text Analytics Practice – NLP Exploration, Ongoing, Python, Sentiment Analysis, Text Classification, Currently developing a natural language processing workflow on anonymised call/chat transcripts to extract sentiment, keywords, and service quality indicators using Python-based models. Processed 5,000+ lines of text so far, achieving 87% sentiment classification accuracy and identifying recurring issues across key experience categories through keyword clustering.