Dontha Harika
Software Engineer @LTM | Gayatri Vidya Parishad College of Engineering (Autonomous)Alumnus|Aspiring MERN Stack Developer| Enthusiastic at Data Science, AIML and Deep Learning
Greater Bengaluru Area
1+ year of work experience
Jul 20263 mosPresent
Bengaluru, Karnataka, India
Apr 20266 mosPresent
Coimbatore, Tamil Nadu, India
Jan 20265 mosMay 2026
Chennai, Tamil Nadu, India
Dec 20252 mosJan 2026
Visakhapatnam, Andhra Pradesh, India
During my internship, I worked on developing and improving a water analysis chatbot as part of the LAMAS project. My work involved building and testing chatbot APIs, fixing response issues, and ensuring that the chatbot gives meaningful and relevant answers instead of directly returning database content. I also worked on debugging the chatbot, improving the quality of cleaned data stored in PostgreSQL, and making the chatbot understand user questions better. In addition, I gained exposure to multi-modal agents, where different data sources and APIs are combined to generate a single, useful response. Overall, the internship helped me understand how real-world chatbots are designed, debugged, and improved to support decision-making in water analysis and sustainability.
Apr 20244 mosJul 2024
Remote
📊 Retail Demand Forecasting Using RapidMiner Studio 🛍️✨
This project focuses on building a robust demand forecasting system for the retail industry using RapidMiner Studio, enabling businesses to optimize inventory management and improve customer satisfaction.
🔑 Key Objectives:
🔹 Demand Prediction: Accurately forecast future demand for products based on historical sales data and market trends.
🔹 Data-Driven Decision-Making: Provide actionable insights to retailers for optimizing stock levels and minimizing overstock or stockout situations.
🔹 Visualization: Present data trends and forecasting results through intuitive dashboards and charts.
🔬 Technical Approach:
Data Preprocessing:
Imported and cleaned historical sales data using RapidMiner workflows.
Handled missing values, outliers, and normalized the dataset for better model accuracy.
Feature Engineering:
Extracted key features like seasonality, holidays, and promotional periods.
Incorporated external factors such as weather and economic indicators where applicable.
Model Building and Evaluation:
Utilized machine learning algorithms (e.g., Linear Regression, Decision Trees, and Random Forest) available in RapidMiner.
Compared multiple models using performance metrics like RMSE, MAE, and R-squared.
Forecasting:
Generated short-term and long-term demand forecasts.
Visualized trends and patterns using RapidMiner’s built-in visualization tools.
🌟 Outcomes:
✅ Improved demand forecast accuracy, reducing inventory mismanagement costs.
✅ Gained insights into seasonal and promotional impacts on product demand.
✅ Enhanced understanding of machine learning workflows in RapidMiner Studio.
🔮 Future Scope:
Incorporating advanced techniques like time-series analysis (ARIMA, Prophet).
Expanding the model to handle dynamic datasets in real-time.
This project demonstrates how data science and predictive analytics can empower the retail industry to make smarter, data-driven decisions! 🛒📈
Dontha’s AI Stack
Education
Bachelors of Technology, Computer Science and Engineering
Gayatri Vidya Parishad College of Engineering (Autonomous)
20214 yrs2025
MPC
Sri Chaitanya College of Education
20192 yrs2021
Join Dontha and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal Boss