Aipower Inc.
Hybrid-Retrieval Sales Chatbot
Dense + sparse retrieval for e-commerce guided selling and troubleshooting.
Jun 2025 – Oct 2025
Framed for
Hybrid dense + sparse retrieval that fixed exact-term failure
Replaced a dense-only baseline with hybrid retrieval (semantic search for intent, keyword search for SKUs and model numbers), improving precision and cutting latency simultaneously.
- Dense + sparse
- hybrid retrieval architecture
- ↓ latency
- versus dense-only baseline
- ↑ precision
- on SKU / model-number queries
- RAG
- Hybrid Retrieval
- LLMs
- Vector Search
- Python
Requirements from the revenue side, translated into an AI solution design
Led technical discovery with marketing and sales, turned their customer requirements into AI solution designs and architecture decisions, and delivered the contracted product ahead of the department wind-down.
- Delivered early
- ahead of department wind-down
- Sales + marketing
- discovery partners
- Discovery-led
- requirements to architecture
- Solution Design
- Technical Discovery
- RAG
- Retrieval Architecture
The problem
Dense-only semantic retrieval failed exactly where e-commerce answers matter most: exact-term queries like SKUs and model numbers, where embedding similarity blurs the very tokens that distinguish one product from another.
What I built
- 01Architected hybrid retrieval combining dense semantic search with sparse keyword search, so exact identifiers stayed exact while natural-language intent still resolved.
- 02Deployed into AI sales chatbots covering both guided selling and product troubleshooting.
- 03Led technical discovery directly with marketing and sales, translating their customer requirements into AI solution designs and architecture decisions.
Outcome
Improved answer precision on exact-term queries and cut response latency versus the dense-only baseline. Delivered the contracted product ahead of the department wind-down.