webInternal ProjectFeatured
Sales Assistant Agent
Demand prediction and promotion optimization platform with LLM agents. Mobile-responsive web application.
Developed at ATP
AI Engineer & Full Stack Developer
2023
Internal project - no public access
Screenshots
About This Project
Developed an intelligent sales assistant that predicts demand and optimizes promotional strategies using LLM agents. The system combines structured POS (Point of Sale) data with unstructured signals like local events, weather patterns, and market trends to provide actionable recommendations for sales teams.
Key Features
- LLM agent for demand prediction and analysis
- Integration of POS data with external signals (events, weather)
- Area-based promotion recommendations
- Real-time inventory optimization suggestions
- Interactive dashboards for sales teams
- Mobile-responsive design for field sales
Challenges & Solutions
- Combining structured and unstructured data sources effectively
- Building reliable demand forecasting models
- Creating actionable recommendations from complex data
- Ensuring real-time performance for sales decisions
Results & Impact
- Improved campaign ROI through optimized promotions
- Better supply planning and reduced stockouts
- Enhanced sales team productivity with AI-powered insights
Technologies Used
ReactPythonLangChainLlamaIndexAWSPostgreSQLRedis
Project Details
- Category
- web
- My Role
- AI Engineer & Full Stack Developer
- Duration
- 6 months
- Year
- 2023
- Company
- ATP
- Status
- Internal / Private
Internal Project
This project was developed for internal use at ATP. Source code and live demo are not publicly available due to confidentiality.