John Li LogoJaison Logo

ARTIFICIAL INTELLIGENCE

Innovative solutions at the intersection of AI, security, and creative technology

Research & Development: AI Systems Innovation

As an active researcher pursuing a Master of Science in Computer Science and AI, I am at the forefront of developing next-generation AI integration methodologies that bridge the gap between cutting-edge artificial intelligence capabilities and existing enterprise infrastructure.

My ongoing R&D work centers on creating intelligent middleware systems that enable seamless communication between modern AI models and legacy enterprise databases. This research addresses a critical industry challenge: how organizations can leverage advanced AI capabilities without completely rebuilding their existing data infrastructure.

My research is contributing to the emerging field of Enterprise AI Infrastructure Engineering - a discipline that didn't exist five years ago but is now critical for organizations seeking to modernize without disruption. Through hands-on implementation with government and educational clients, I'm developing practical frameworks that other organizations can adopt for their own AI transformation initiatives.

Intelligent Document Ecosystem

Active
2025

A suite of interconnected projects forming the Intelligent Document Ecosystem. Core objective is to professionalize and expand document processing, analysis, and information retrieval using advanced AI, particularly Large Language Models (LLMs), and robust data management.

Key Features

  • Structured metadata extraction
  • Intelligent data filtering
  • Robust LLM integration
  • Complex document understanding

Technologies

PythonLLMsData EngineeringAI/ML
Private ProjectCode

Firewood

Completed
2025

Professional Chat & Document Processing Toolkit. A comprehensive, production-ready toolkit for processing structured and unstructured documents programmatically and with AI-powered analysis.

Key Features

  • Multi-format parsing (ChatGPT, Claude, Codex)
  • Hybrid regex + LLM cleaning approaches
  • AI-powered analysis and topic extraction
  • Massively parallel processing with intelligent batching

Technologies

PythonOpenAI GPTGoogle GeminiParallel Processing
Private ProjectCode

tXt-ray

Completed
2025

A simple but powerful browser-based text comparison tool with Git-like features, inline comments, blame view, and advanced search capabilities.

Key Features

  • Multiple view modes (Git Unified, Overlay, Side-by-Side)
  • Word-level diff detection
  • Smart chunking with context
  • Comprehensive export options (.patch, text files)

Technologies

JavaScriptBrowser-basedGit IntegrationWeb Tools
Private ProjectCode

Dupstep

Completed
2025

Document Duplicate Detection Tool. A powerful Python tool that analyzes documents to find and report duplicate content using multiple detection methods including exact matching, fuzzy matching, and AI-powered semantic similarity.

Key Features

  • Multiple detection methods (exact, fuzzy, semantic)
  • Beautiful HTML reports with color-coding
  • Comprehensive export options (CSV, Excel, JSON)
  • AI-powered semantic similarity using embeddings

Technologies

PythonFuzzyWuzzySentence-BERTAI Embeddings
Private ProjectCode

Seymour

Completed
2025

Computer vision project involving vision machine learning algorithms. Volume estimation monitoring via Reolink E1 Pro camera - archiveing data for analysis.

Key Features

  • Icon-based volume detection using Computer Vision
  • Automatic media playback control via webhooks
  • Conditional LLM anomaly detection
  • YOLOv8 segmentation model with training toolkit

Technologies

PythonYOLOv8Computer VisionRTSPLLM Integration
Private ProjectCodeView Research

LLM Whisperer

Completed
2024

PDF Processing Pipeline. A robust PDF processing pipeline using LLMWhisperer for high-quality text extraction with layout preservation.

Key Features

  • High-quality PDF text extraction
  • Layout preservation
  • Robust processing pipeline
  • API integration

Technologies

PythonLLMWhisperer APIPDF Processing

Repo-rt

Completed
2025

Transform your code repositories into actionable insights for both human understanding and advanced AI analysis. A powerful Python-based toolkit designed to scan local repositories or directories.

Key Features

  • Comprehensive artifact generation
  • Intelligent content prioritization
  • Interactive HTML for human browsing
  • Structured data optimized for AI

Technologies

PythonRepository AnalysisAI IntegrationHTML/JSON

Tuner

Completed
2024

A comprehensive pipeline for preparing, cleaning, and optimizing content for fine-tuning language models on your writing style and tone. Takes content from various sources and processes it through several stages.

Key Features

  • Content extraction from multiple sources
  • Text cleaning and preprocessing
  • AI-driven command extraction/refinement
  • Dataset creation for various training frameworks

Technologies

PythonOpenAI APIContent ProcessingML Pipeline

School Safety Dashboard

In Progress
2024

A comprehensive safety monitoring and compliance system for educational institutions with real-time reporting and incident management.

Key Features

  • Real-time safety monitoring
  • Incident reporting system
  • Compliance tracking
  • Parent/guardian notifications

Technologies

ReactNode.jsMongoDBFirebase
Private ProjectCode

Digital Asset Management

Completed
2024

A comprehensive digital asset management system for creative teams with AI-powered tagging and search capabilities.

Key Features

  • AI-powered auto-tagging
  • Advanced search and filtering
  • Version control system
  • Team collaboration tools

Technologies

ReactExpress.jsMongoDBAWS S3
Private ProjectCode

Pubwish

Completed
2019

A writing platform app that helps writers track progress, collaborate, and improve their writing habits through gamification and social interaction. The app combines powerful writing tools with social features and challenges to encourage consistent writing habits.

Key Features

  • Authentication with multiple social login options
  • Document editing and writing management
  • Writing challenges and competitions
  • Social features (friends, chat, groups)

Technologies

iOSSwiftXcodeCocoaPods
Private ProjectCode

Get the latest updates
direct to inbox