Shubhashish Prakash - Associate Lead Analyst - Concentrix

Summary

Data Scientist with 3.5+ years of experience driving demand forecasting, predictive modelling, workforce capacity optimisation, and data-driven planning strategyacross large-scale operational environments. Proven track record of improving forecast accuracy, reducing operating costs, and enabling risk-aware decision making through advanced statistical modelling, machine learning, and probabilistic forecasting frameworks. Experienced in building scalable analytics automation solutions and applied Generative AI workflows that enhance executive visibility, operational efficiency, and strategic workforce planning outcomes.

Overview

4

years of professional experience

1

Certification

Work History

Associate Lead Analyst

Concentrix

12.2024 - Current

Led large-scale time-series demand forecasting and predictive analytics initiatives on 10+ years of high-volume toll transaction data across 8 Lines of Business, improving demand visibility and strengthening data-driven operational planning.
Designed and deployed multi-horizon forecasting models (interval, daily, weekly, monthly) using statistical and machine learning techniques while incorporating external demand drivers such as weather variability, holidays, traffic seasonality, and government shutdown events, improving overall forecast accuracy by ~15%.
Built enterprise-level workforce capacity planning and headcount optimisation frameworks integrating demand forecasts, shrinkage modelling, service level targets, and scenario simulations, enabling risk-aware staffing strategies and contributing to ~10% reduction in operating costs.
Implemented reproducible forecasting workflows and model validation pipelines to standardise demand prediction processes, improve model reliability, and support scalable analytics operations.
Developed automated KPI monitoring dashboards and reporting pipelines supporting forecast performance tracking across 8 Lines of Business, improving executive visibility into demand variability, reducing manual reporting effort, and accelerating data-driven decision making.
Conducted advanced data pre-processing, feature engineering, exploratory analysis, and trend decomposition to support predictive modelling, probabilistic forecasting initiatives, and demand variability assessment.
Collaborated with senior operations leaders, workforce management teams, and cross-functional stakeholders to translate forecasting insights into actionable staffing and resource allocation strategies, improving utilization efficiency and workforce productivity.

Data Analyst

Oneness Buildtech Pvt Ltd

08.2022 - 12.2024

Engineered scalable data cleaning, validation, and transformation pipelinesacross multi-source structured datasets, improving data integrity and enabling reliable predictive analytics and demand forecasting workflows.
Developed statistical forecasting modelsand regression-based analytical solutions to enhance demand planning visibility, identify sales and operational trends, and support inventory and resource optimisation.
Performed cross-functional data analysis with finance and marketing teams to evaluate performance drivers, campaign effectiveness, and cost-efficiency opportunities through exploratory analytics and trend diagnostics.
Automated recurring business intelligence reporting processes and performance dashboards using Python and Excel-based analytics tools, significantly improving analytical productivity and reporting efficiency.

Education

Bachelor of Technology (B. Tech) - Civil Engineering

Dr. APJ Abdul Kalam Technical University

Lucknow

07-2018

Bachelor of Laws (LLB) -

Lucknow University

06-2021

Skills

Programming & Data: Python, SQL, Advanced Excel, Jupyter
Machine Learning: Linear & Logistic Regression, Ridge/Lasso, Decision Trees, Random Forest, Gradient Boosting, XGBoost, KNN
Model Development: Feature Engineering, Cross-Validation, Hyperparameter Tuning
Time-Series & Forecasting: ARIMA, SARIMA/SARIMAX, Holt-Winters, ETS, Prophet, TBATS, Multi-Horizon Forecasting, Probabilistic Forecasting, Quantile Regression, Prediction Intervals
Model Evaluation: MAPE, RMSE, MAE, Forecast Bias

Forecast Backtesting & Model Monitoring:
Rolling validation, Error tracking, Model performance drift analysis
Deep Learning: ANN, RNN, LSTM (Time-Series), TensorFlow / Keras
Data Science Stack: Pandas, NumPy, Scikit-learn, Statsmodels, Matplotlib, Seaborn
Generative AI: Prompt Engineering, Conversational AI, Retrieval-Augmented Generation (RAG), LLM-Driven KPI Automation
Domain Expertise: Demand Forecasting, Workforce & Capacity Planning, Scenario Modelling, Market Mix Modelling, Churn & Business Analytics

Certification

• Microsoft Generative AI Certificate
• Data Science – AnalytixLabs

Projects

LLM-Based Business Analytics Assistant

Tools: Python, Pandas, OpenAI API, Prompt Engineering.
Engineered a GPT-powered analytics assistant integrating retrieval-augmented generation (RAG) workflows to automate KPI interpretation, business query resolution, and executive report summarisation.
Designed retrieval-augmented pipelines integrating structured datasets and document knowledge bases using embedding-driven similarity search for context-aware analytics responses.
Developed prompt orchestration workflowsfor automated KPI narrative generation, anomaly explanation, and trend summarisation, reducing manual analytical effort.
Integrated forecasting outputs and regression-based scenario simulations into conversational workflows to support workforce planning and decision support use-cases.

Probabilistic Demand Forecasting & Capacity Risk Optimization

Tools: Python, Scikit-learn, Statsmodels, Prophet, TBATS
Developed probabilistic multi-horizon demand forecasting models generating prediction intervals to support uncertainty-aware workforce and capacity planning.
Modelled complex multi-seasonal demand behaviour using TBATS and quantile regression techniques to improve forecast robustness for high-frequency operational datasets.
Built scenario-based simulation frameworks to evaluate staffing risk across peak, baseline, and low-demand conditions, strengthening planning confidence and cost optimisation decisions.
Evaluated forecast reliability using quantile loss and interval coverage metrics and delivered visual risk-band dashboards for leadership decision support.

Production-Grade Forecasting & Workforce Optimization

Tools: Python, Prophet, Excel, Python, FastAPI, LightGBM, Airflow, Docker, Streamlit
Designed and deployed an end-to-end forecasting pipeline automating data ingestion, feature engineering, model training, and probabilistic prediction generation.
Built REST-based forecasting service enabling real-time demand prediction and staffing scenario simulation.
Implemented scheduled retraining workflows and model versioning to ensure forecast reliability and scalability.
Developed optimisation engine to recommend workforce allocation strategies under demand uncertainty, improving simulated cost efficiency and service level adherence.
Created interactive dashboard visualizing forecast trends, prediction intervals, and capacity risk exposure for decision support.

Publications And Leadership

Publications: Research work published in a globally recognized international journal under my authorship.
Data Surveying: Volunteered in extensive village surveys under the Uttar Pradesh Rajya Nirmaan Sahkari Sangh (UPRNSS).
Community Leadership: Volunteered at Rotary Blood Bank, Ghaziabad. Served as General Secretary of the College Cultural Club and Coordinator for the AKTU State Level Sports Fest

Timeline