About Me
Hi, I'm John Carroll—a passionate software engineer and technologist. I specialize in building modern, scalable web applications and AI-powered solutions. My expertise spans full-stack development, cloud architecture, and developer experience, with a focus on delivering impactful products and delightful user experiences.
I enjoy working with cutting-edge technologies, collaborating with diverse teams, and sharing my knowledge through talks and blog posts. Whether it's architecting robust backends, crafting beautiful frontends, or exploring the latest in AI, I thrive on solving complex problems and learning new things.
Let's connect if you'd like to collaborate, discuss technology, or just say hello!
Projects
Portfolio
Started with a Vercel template and made it my own! Personal portfolio site showcasing projects and skills.
Extract Image from Product URL
Demonstrates how to use OpenAI's Response API (with GPT-4.1 and tool calling) to extract the main product image URL from an e-commerce product page. Provides both Python and TypeScript implementations.
Supply Chain Reasoning Engine
Adaptive Supply Chain Optimization System using AI reasoning models for real-time disruption response and resilience.
Field Technician Assistant
Multimodal RAG with voice interface for field technicians. Uses Azure AI Search & OpenAI for hands-free access to technical documentation through speech and image analysis.
Write Customer Success Stories
RAPID: Real AI-Powered Insights on Demand. Centralizes and automates the creation of customer success stories using data pipelines and GPT-4 AI, with AI-powered insights and end-to-end automation.
OpenAI Chat Completions Function Calling Examples
A collection of Python scripts demonstrating the power and flexibility of the Chat Completions API, focusing on function calling, prompt suggestions, conversation summarization, and streaming responses.
Coffee Chat Voice Assistant
Voice-driven ordering system powered by Azure OpenAI GPT-4o Realtime API. Simulates a café barista experience with natural conversations, live order updates, and real-time transcription. Showcases AI for seamless customer interactions.
Video Analysis with LLMs
A Streamlit app leveraging AI vision to analyze video content, extract insights, and enable interactive conversations about visual content. Supports video upload, segmentation, and chat about results.
Image2Csv
A Streamlit app that leverages Azure OpenAI's vision capabilities to convert images containing tables or structured data into CSV format for easy use in spreadsheets and data analysis.
Extract Content from Documents
Implements field extractors that convert PDF documents to JSON using OpenAI’s GPT-4 Vision Preview. Supports scanned and digital PDFs, batch processing, and schema-based extraction.
Product Categorization / Classification
Automates product categorization using Generative AI models. Leverages Azure AI Search and OpenAI's GPT-4 to classify products based on their descriptions and other attributes.
Azure OpenAI Search Demo
A RAG chat app that uses Azure OpenAI Service and Azure AI Search to create ChatGPT-like experiences over your own data. Features multi-turn chat, Q&A, citations, and support for many document formats.
AI Tutor
AI Tutor POC that identifies the source of student misunderstanding, using quizzes and targeted review to help students master difficult concepts.
Planogram Analysis
Uses Azure OpenAI GPT-4o to audit retail planograms and product assortments. The AI auditor inspects each shelf of a cooler photo, determining if it complies with the planogram rules.
Job Description Match to Resumes
JobFit AI Matchmaker is a proof of concept that reimagines resume-to-job matching using advanced Azure AI services. Matches resumes to a user-provided job description with high accuracy and relevance.
Voice Assistant using GPT-4o Realtime
Voice Assistant using GPT-4o Realtime. Provides a hands-free, voice-driven experience for various use cases.
OpenAI Console App with Logging per Image Processed
A console app that uses GPT-4o for object detection in images, logging each image processed, its response, and end-to-end request time.
Multi-Agent Example: Semantic Kernel and LLM's to Execute Python Code
Demonstrates using Semantic Kernel and LLMs to execute Python code in a sandboxed environment. Features safe code execution, logging, and agent group chat.
Computer Use Agent
An intelligent agent designed to automate and optimize business computer usage workflows. Features include task automation, user activity monitoring, and integration with enterprise systems. (Private project, details coming soon.)
Multi-Agent Network and registration
A platform for registering, managing, and orchestrating multiple AI agents in a networked environment. Enables agent discovery, communication, and collaboration for complex workflows. (Private project, details coming soon.)
Speech I/O Agent
A reusable agent for speech input and output, supporting Google A2A, agent registration, and discoverability. Enables natural language voice interactions and seamless integration with other AI agents. (Private project, details coming soon.)
Tech Stack
Career Timeline
Coming soon: A visual timeline of my career journey, roles, and key milestones.
Speaking Engagements
Talks, workshops, and panels at conferences and meetups will be listed here.
- [Conference/Workshop Placeholder]
Blogs Published
A curated list of my published blog posts and articles will appear here.
- [Blog Post Placeholder]
Stealth Startup Work
Details about my work on stealth and early-stage startups will be shared here (as much as I can reveal!).