Design a Hybrid Symbolic-Neural Agent for an Enterprise RAG Demo
Overview
What this challenge is about.
Design a hybrid agent for a 'company-policy assistant' demo: a symbolic planner decomposes user goals into typed subtasks ('find policy', 'check applicability', 'compose answer'), then a RAG layer (vector store + LLM) executes the retrieval subtasks. Build a working prototype on around 30 anonymized policy documents, demonstrate 5 sample queries, and instrument latency and citation accuracy. Produce a 30-slide deck plus a 2-page methodology rationale arguing where the symbolic layer earns its keep vs. a pure-LLM baseline. Workshop must feel polished and run in under 60 minutes total.
The Brief
What you'll do, and what you'll demonstrate.
Build and present a hybrid symbolic-neural enterprise assistant that beats a pure-LLM baseline on citation accuracy.
Earning criteria — what you'll demonstrate
- Design a hybrid symbolic-plus-neural architecture for a realistic task
- Implement a small planner and integrate it with a RAG back-end
- Measure citation accuracy as a deployable-LLM metric
- Present a methodology argument to a skeptical enterprise audience
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Solutions Architect
Designing a hybrid architecture that respects both classical and neural AI is the architect's daily craft inside enterprise consultancies.
This challenge sharpens
- hybrid-ai
- symbolic-planning
- retrieval-augmented-generation
AI Engineer
Wiring up a planner plus RAG back-end into a working demo is the AI engineer's bread and butter at any AI-product company.
This challenge sharpens
- python
- retrieval-augmented-generation
- symbolic-planning
Prompt Engineer
Tuning the LLM layer for citation accuracy and instrumenting it for evaluation is precisely the prompt engineer's day-one work.
This challenge sharpens
- llm-evaluation
- retrieval-augmented-generation
- stakeholder-communication