Personal AI Advisor
A personal AI advisor that answers questions about skills, experience, and services using structured data and website content.
Project Overview
The Personal AI Advisor is a WordPress-based AI system designed to act as an intelligent assistant representing a professional’s skills, experience, and services. The advisor can answer questions from recruiters, hiring managers, and potential clients by leveraging structured data feeds, website content, and custom knowledge sources. The system connects a frontend chat interface with a Cloudflare Worker acting as an AI gateway, enabling dynamic retrieval and injection of contextual knowledge into LLM responses. The goal of the project was to create a scalable and controllable AI representation layer that can communicate professional value clearly and consistently without manual intervention. This project demonstrates my ability to build AI-driven communication tools, design knowledge-driven systems, and orchestrate LLM-based architectures for real-world use cases.
My Role
I designed and built the full system, including the WordPress plugin interface, knowledge ingestion pipeline for structured and unstructured data, Cloudflare Worker AI gateway, prompt design and guardrails, and the frontend chat experience. The project required combining AI orchestration, data structuring, and UX design to create a reliable and context-aware personal assistant.
The Problem Context
Professionals often struggle to communicate their full range of skills, experience, and capabilities effectively across different audiences such as recruiters, hiring managers, and clients.
- Information about skills and experience is scattered across resumes, websites, and documents.
- Recruiters and clients cannot easily explore capabilities in an interactive way.
- Static content does not adapt to different questions or contexts.
- Generic AI chatbots lack personalized and structured knowledge.
The objective was to create an AI advisor capable of answering questions dynamically while being grounded in accurate, structured personal data.
Key Features & Workflow
Context-Aware AI Advisor
Answers questions about skills, experience, and services using personalized knowledge sources.
Multi-Source Knowledge Ingestion
Uses structured data feeds (CSV, JSON), website pages, and text files as contextual input.
Interactive Chat Interface
Embeddable chat UI allowing recruiters and clients to explore capabilities in real time.
AI Guardrails & Prompt Control
Ensures responses remain accurate, relevant, and aligned with the professionalu2019s profile.
Technical Architecture
How the system components fit together.
WordPress Chat Interface
Frontend component embedded on the website to handle user interaction and messaging.
Cloudflare Worker Gateway
Manages LLM requests, injects knowledge context, and handles prompt orchestration.
Knowledge Sources
Includes structured data feeds, website pages, and external documents representing skills and experience.
Design Decisions
A key design principle was to ensure the AI provides accurate and controlled responses grounded in real data.
- Ingest structured and unstructured data sources.
- Process and organize knowledge for AI consumption.
- Inject relevant context into each user query.
- Generate accurate, contextual responses via the LLM.
This approach transforms static professional information into a dynamic, interactive experience for recruiters and clients.
Future Roadmap
- Semantic search using vector embeddings.
- Integration with CRM and recruiting platforms.
- Personalization based on user type (recruiter vs client).
- Analytics on user interactions and questions.
- Voice-based AI advisor interface.
Technology Stack
Core technologies used to build this project.
Capabilities Demonstrated
Key areas of expertise highlighted by this project.
