Portfolio
The work.
28 projects across AI, fintech, e-commerce, computer vision, blockchain, and enterprise systems — each with the real challenge, the real solution, and the techniques that made it ship.
This is a curated selection. Many projects are under NDA and can't be listed here, and some older work predates this portfolio. The full experience timeline is on the resume page.
NeuralBridge
16 PROJECTS
Enterprise-grade AI expert matching platform connecting organizations with pre-vetted AI specialists. LangChain-orchestrated multi-model evaluation pipeline (LLaMA, Mistral, GPT, Gemini) for parallel expert scoring across technical depth, domain specificity, and communication clarity. Hybrid fraud detection (LLM + ML anomaly detection) and RAG-powered semantic matching using vector embeddings.
CHALLENGE
Trust and relevance at scale — how do you verify that an "AI expert" is actually an expert, and match them to the right project without manual review?
SOLUTION
Built a LangChain-orchestrated pipeline where 4 models (LLaMA, Mistral, GPT, Gemini) evaluate each candidate in parallel across different dimensions. Hybrid fraud detection layers LLM semantic analysis with ML anomaly detection to catch inconsistencies rule-based systems miss. RAG-powered matching via Pinecone/Weaviate replaced keyword search entirely.
TECHNIQUES
- →Multi-model ensemble scoring
- →LangChain chain orchestration
- →RAG with Pinecone & Weaviate
- →Hybrid fraud detection (LLM + ML)
- →Parallel async evaluation pipelines

MR Pay
Digital payment platform similar to Alipay. Designed secure transaction flows, multi-party settlement logic, and scalable backend services for web and mobile. Blockchain-inspired transaction signing for payment security.
CHALLENGE
Building a payment platform at Alipay scale — handling secure transactions, multi-party settlement, and preventing double-spending without a distributed ledger.
SOLUTION
Designed blockchain-inspired transaction signing where every transaction gets a cryptographic signature. Multi-party settlement logic handles vendor splits, refunds, and escrow. Built scalable backend services supporting both web and mobile clients.
TECHNIQUES
- →Blockchain-inspired transaction signing
- →Multi-party settlement logic
- →Cryptographic transaction verification
- →Scalable payment architecture

MR Bay
Full-scale e-commerce ecosystem comparable to Alibaba. Order processing, vendor management, inventory, and payment integration across multiple storefronts.
CHALLENGE
Building an e-commerce platform at Alibaba scale — order processing, vendor management, inventory tracking, and payment integration across multiple storefronts with real traffic.
SOLUTION
Architected a full-stack ecosystem with React frontend, Node.js backend, and MongoDB. Integrated MR Pay for payment processing. Every component had to handle real traffic at real volume — no staging environment luxury.
TECHNIQUES
- →Multi-storefront architecture
- →Real-time inventory management
- →Vendor management system
- →Payment integration with MR Pay

StrongGate
Computer vision pipeline processing 280 Hikvision camera feeds to control physical access gates via PLC integration. YOLO-based object detection and OpenCV image processing for license plate recognition. Also built a thermal-imaging attendance system during COVID.
CHALLENGE
280 Hikvision cameras feeding live video to control physical access gates. Every gate open/close cycle had to be reliable — software failures meant real vehicles stuck at real gates. License plate recognition had to work in varying lighting, weather, and angles.
SOLUTION
Built a YOLO-based object detection pipeline for vehicle detection, OpenCV for license plate recognition, and face detection for driver verification. PLC integration handles physical gate control with fail-safe defaults. Thermal imaging extension during COVID added body temperature screening on top of access control.
TECHNIQUES
- →YOLO object detection for vehicles
- →OpenCV license plate recognition
- →Face detection for driver verification
- →PLC gate control with fail-safe
- →Thermal imaging anomaly detection

AI book writing assistant with multi-stage prompt pipelines (outline generation → structured drafting → revision loops) using GPT and Gemini. Maintains narrative consistency across chapters through context window management and stateful prompt chaining.
CHALLENGE
AI-generated long-form content loses narrative consistency across chapters. Context windows can't hold an entire book, so earlier plot points get forgotten, characters drift, and themes fragment.
SOLUTION
Multi-stage prompt pipeline: outline generation first, then structured drafting chapter-by-chapter, then revision loops. Stateful prompt chaining carries key narrative elements (character states, plot threads, themes) forward across chapters. Context window management ensures the AI always has the right context at the right stage.
TECHNIQUES
- →Multi-stage prompt pipelines (outline → draft → revise)
- →Stateful prompt chaining across chapters
- →Context window management
- →GPT + Gemini dual-model workflow
- →WordPress integration for publishing

AI audiobook generation platform. Authors load books as text files, assign character voices to dialogue lines, and automatically generate complete audiobooks. Integrates ElevenLabs APIs with WordPress.
CHALLENGE
Converting books to audiobooks is expensive and slow. Authors need a way to assign different voices to different characters, handle narration vs dialogue, and generate hours of audio without manual recording.
SOLUTION
Built a platform where authors upload text files, map character names to ElevenLabs voices, and the system automatically detects dialogue lines and assigns the right voice. Narration gets a separate narrator voice. WordPress integration handles publishing and distribution.
TECHNIQUES
- →ElevenLabs API integration for voice synthesis
- →Dialogue detection and character voice mapping
- →Automated audio generation pipeline
- →WordPress publishing integration

Real-time cryptocurrency analytics platform (BubbleMap-style) with sub-second update latency on on-chain data streams. Unsupervised anomaly detection flags unusual wallet clustering and transaction velocity patterns.
CHALLENGE
On-chain data moves fast — thousands of transactions per second across multiple chains. Visualizing this in real-time with sub-second latency while also detecting anomalies requires a carefully engineered pipeline.
SOLUTION
WebSocket streaming pipeline with Node.js backend processing live on-chain data. D3.js-powered BubbleMap visualization renders wallet clusters and transaction flows in real-time. Unsupervised anomaly detection runs alongside, flagging unusual wallet clustering and transaction velocity patterns invisible in standard charting tools.
TECHNIQUES
- →WebSocket streaming for sub-second updates
- →Unsupervised anomaly detection on transaction patterns
- →D3.js BubbleMap visualization
- →Real-time on-chain data processing
AI character generation and chat platform with image generation (pose/style controls) and conversational AI companions. Powered by SPICE AI tools, currently at v4.9.
CHALLENGE
Building an AI platform that generates consistent character images across different poses and styles while maintaining coherent multi-turn conversations.
SOLUTION
Integrated SPICE AI tools for image generation with pose and style controls. Conversational AI companions maintain context across sessions. Scaled to v4.9 with active users.
TECHNIQUES
- →SPICE AI image generation
- →Pose and style control pipelines
- →Multi-turn conversational AI
- →Character consistency across generations

MindAI SwapBot
LLM-driven Telegram bot that interprets natural language swap requests and executes on-chain transactions. Intent-parsing pipelines handle ambiguous user inputs gracefully — critical for financial actions where misinterpretation has real consequences.
CHALLENGE
Users describe swap requests in natural language — "swap 100 USDC for ETH" — but phrasing varies wildly. Misinterpreting a financial transaction has real monetary consequences.
SOLUTION
LLM-driven intent parsing pipeline that normalizes natural language into structured swap parameters. Ambiguous inputs trigger clarification prompts rather than executing wrong trades. On-chain transaction execution with confirmation callbacks.
TECHNIQUES
- →LLM intent parsing for natural language swaps
- →Ambiguity detection with clarification prompts
- →On-chain transaction execution
- →Telegram bot integration

Relationship and marriage counseling platform with content delivery and subscription workflows. The nine-year project — started on Upwork, survived two accidents, and finally delivered.
CHALLENGE
A nine-year journey — started on Upwork, survived two car accidents, and finally delivered. The client needed a counseling platform with content delivery, subscription management, and a seamless user experience.
SOLUTION
Built a full-stack platform with React frontend, Node.js backend, subscription workflows, and content delivery pipelines. The project taught patience and persistence as much as engineering.
TECHNIQUES
- →Subscription workflow management
- →Content delivery pipelines
- →React + Node.js full-stack architecture

MR USBLock
Digital asset access control system with ML-based behavioral anomaly detection. Flags access pattern deviations before they escalate to breaches.
CHALLENGE
Traditional access control only checks credentials — but what if an authorized user starts behaving abnormally? Detecting insider threats requires behavioral analysis, not just authentication.
SOLUTION
ML-based behavioral anomaly detection monitors access patterns in real-time. Deviations from established baselines trigger alerts before they escalate to breaches. Layered on top of standard access control for defense in depth.
TECHNIQUES
- →ML-based behavioral anomaly detection
- →Real-time access pattern monitoring
- →Baseline deviation alerting
- →Defense-in-depth access control

BRC-20 inscription platform and airdrop point system. Built frontend platform and website for blockchain inscription services.
CHALLENGE
BRC-20 inscription on Bitcoin requires specialized tooling. Users need a clean interface to create inscriptions, track them, and participate in airdrop point systems.
SOLUTION
Built the frontend platform and website for MemoLabs — handling BRC-20 inscription creation, tracking, and an airdrop point system for community engagement.
TECHNIQUES
- →BRC-20 inscription interface
- →Airdrop point system
- →Web3 frontend integration
Drive3
Web3 decentralized storage platform for secure, distributed file storage on the blockchain.
CHALLENGE
Centralized storage has single points of failure. Building a Web3 storage platform that distributes files across a decentralized network while maintaining reliability and speed.
SOLUTION
Built a Web3 decentralized storage platform leveraging blockchain for file integrity verification and distributed nodes for storage redundancy.
TECHNIQUES
- →Decentralized file storage
- →Blockchain integrity verification
- →Distributed node architecture

Reshoper
Enterprise-grade Restaurant & Shopping Mall Management Platform covering web and app interfaces. Desktop-based restaurant management software with POS integration.
CHALLENGE
Restaurants and shopping malls need unified management — POS, inventory, staff scheduling, reporting — across both web and desktop interfaces. Different environments, same data.
SOLUTION
Delivered enterprise-grade management platform with web and app interfaces plus desktop-based restaurant management software. POS integration handles real-time transactions and inventory updates.
TECHNIQUES
- →Cross-platform architecture (web + desktop)
- →POS integration
- →Real-time inventory and staff management
- →Enterprise reporting dashboards
Lottery Platform
High-concurrency lottery platform built in Rust with Svelte frontend and PostgreSQL. Handles 100,000+ simultaneous ticket purchases with sub-second latency. Transactional integrity is non-negotiable — every ticket must be atomic, every draw must be verifiable, and every user must see consistent state.
CHALLENGE
100,000 users buying lottery tickets simultaneously creates extreme write contention on the database. Every ticket purchase must be atomic — no double-sales, no lost tickets, no race conditions. Traditional web frameworks buckle under this load, and a single failed transaction means a real user loses money.
SOLUTION
Built the backend in Rust for zero-cost abstractions and predictable memory management. PostgreSQL handles transactional integrity with row-level locking and SERIALIZABLE isolation for ticket sales. Connection pooling with bounded async tasks prevents resource exhaustion under peak load. Svelte frontend delivers a lightweight client-side experience with real-time WebSocket updates for draw results and ticket status.
TECHNIQUES
- →Rust async runtime for high-concurrency handling
- →PostgreSQL row-level locking with SERIALIZABLE isolation
- →Bounded connection pooling to prevent resource exhaustion
- →WebSocket real-time draw updates
- →Atomic ticket purchase transactions
- →Svelte reactive frontend with minimal bundle size

Contact Aggregation System
Contact information aggregation platform combining automated scraping, validation, and deduplication pipelines. Designed for large-scale outreach data collection with scheduling and export workflows.
CHALLENGE
Collecting contact data across thousands of heterogeneous sources without getting blocked, while keeping the resulting dataset clean enough to actually use.
SOLUTION
Built rotating, rate-limited scraping workers with headless-browser fallbacks for JavaScript-heavy sites, then ran every record through validation and fuzzy deduplication before export.
TECHNIQUES
- →Headless-browser scraping with rate limiting
- →Validation and fuzzy deduplication pipelines
- →Scheduled collection jobs with retry logic
NovaCircuit
12 PROJECTS
MR Eye
Endpoint monitoring solution with screen recording, keystroke tracking, and behavioral anomaly detection for enterprise security oversight.
CHALLENGE
Enterprise security requires monitoring endpoints without blocking productivity. The system had to capture screen activity, track keystrokes, and detect anomalous behavior — all without noticeable performance impact.
SOLUTION
Built a lightweight endpoint agent with screen recording, keystroke tracking, and behavioral anomaly detection. The anomaly detection engine flags unusual patterns — mass file copies, off-hours access, unexpected application usage — for security review.
TECHNIQUES
- →Low-overhead screen recording
- →Keystroke pattern tracking
- →Behavioral anomaly detection engine
- →Real-time security alerting

MeetUp
Windows desktop video conferencing application with real-time audio/video streaming. Also built a web-based version using Node.js, MongoDB, and WebRTC.
CHALLENGE
Real-time video conferencing on Windows desktop with multi-party audio/video streaming. Latency, synchronization, and bandwidth management are all critical.
SOLUTION
Built a Windows desktop application with real-time audio/video streaming. Later extended to a web-based version using Node.js, MongoDB, and WebRTC for browser-based communication.
TECHNIQUES
- →Real-time audio/video streaming
- →WebRTC for browser-based conferencing
- →Multi-party synchronization
- →Bandwidth-adaptive streaming

TSolution
Office suite built on top of LibreOffice. Served as PM and led development of the HWP 5.0 document parsing module and the main UI. The printed LibreOffice class and module documentation covered three full office walls.
CHALLENGE
HWP is a proprietary Korean document format with sparse documentation, and LibreOffice’s internal module architecture is famously vast — the printed class documentation covered three office walls.
SOLUTION
Reverse-engineered the HWP 5.0 binary format against the published spec, mapped it onto LibreOffice’s document model through the UNO API, and led the module and main UI development as PM.
TECHNIQUES
- →HWP 5.0 binary format parsing
- →LibreOffice UNO API integration
- →Document model mapping
- →Team leadership as PM

KakaoTalk Desktop Client
Desktop chat application for KakaoTalk, Korea’s dominant messenger, with messaging automation capabilities built on top of protocol analysis.
CHALLENGE
Building a desktop client and automation layer for a closed, proprietary messaging protocol meant reverse-engineering the wire format and keeping up with server-side changes without any official API.
SOLUTION
Analyzed the KakaoTalk protocol through packet inspection, implemented a compatible desktop client in C# on Windows, and layered messaging automation on top with graceful handling of protocol updates.
TECHNIQUES
- →Protocol reverse engineering via packet analysis
- →Windows desktop client in C#
- →Messaging automation layer
- →Resilience to server-side protocol changes

ArchDic
Offline dictionary application of architectural terms for architects. What looked like a simple lookup tool turned out to rest on corpus theory at its core — the first project that taught how linguistic data structures really work.
CHALLENGE
A dictionary app sounds trivial until you need fast offline lookup over a large domain-specific corpus with fuzzy matching — corpus theory turned out to be at the heart of it.
SOLUTION
Built an indexed offline dictionary in C++/Qt with prefix and fuzzy search over a curated architectural-terms corpus, packaged for desktop with no network dependency.
TECHNIQUES
- →Corpus-based dictionary construction
- →Indexed offline full-text search
- →Prefix and fuzzy matching
- →Qt desktop packaging

3D Modeling Tool (Revit)
A 3D modeling tool integrated with the Autodesk Revit SDK, built during the KOAST internship. The project was the first hands-on lesson in working inside a large third-party SDK and its object model.
CHALLENGE
Integrating with the Autodesk Revit SDK meant learning a massive, opinionated BIM object model and API surface, then bending it toward custom 3D modeling behavior.
SOLUTION
Studied the Revit SDK object model, built a modeling tool on top of its .NET API, and mapped custom modeling operations onto Revit’s document and element structures.
TECHNIQUES
- →Autodesk Revit SDK / .NET API integration
- →BIM object-model mapping
- →Custom 3D modeling operations

Cocos2d Game Ports
Ported two SWF Flash games (Mortal Combat, Fighter Flight) to Cocos2d for Android and iOS during the KOAST internship — the assignment that followed the legendary Tetris code-golf competition.
CHALLENGE
Reimplementing Flash/SWF games on Cocos2d for both Android and iOS meant rebuilding game logic and asset pipelines that originally targeted the Flash runtime, on two mobile platforms at once.
SOLUTION
Recreated the games’ mechanics and scene flow in Cocos2d, rebuilt the asset and animation pipeline for mobile, and shipped to both Android and iOS from a shared C++ codebase.
TECHNIQUES
- →SWF-to-Cocos2d game reimplementation
- →Shared C++ codebase for Android + iOS
- →Mobile asset and animation pipeline

CAPTCHA Recognition
A CAPTCHA auto-recognition and bypass system built right as TensorFlow and Keras were first emerging. Combined classic image processing with early deep-learning models to read distorted challenge text.
CHALLENGE
Reading deliberately distorted CAPTCHA text required tolerance to noise, warping, and overlapping characters — well beyond what template matching could handle, at a time when deep-learning tooling was brand new.
SOLUTION
Built an image-processing front end (denoise, segment, normalize) feeding CNN models trained in TensorFlow/Keras to recognize characters, iterating on architectures as the frameworks matured.
TECHNIQUES
- →Image preprocessing and character segmentation
- →CNN-based character recognition (TensorFlow/Keras)
- →Synthetic training-data generation
- →End-to-end recognition pipeline

APK Reverse Engineering
Android APK and Unity APK reverse engineering — analyzing native Android apps and Unity builds (4.x–5.6) to understand, audit, and rebuild their behavior.
CHALLENGE
Understanding shipped Android and Unity applications with no source meant decompiling, deobfuscating, and reconstructing program behavior across both Dalvik bytecode and Unity’s IL2CPP/Mono internals.
SOLUTION
Reverse-engineered APKs through smali/DEX analysis and native inspection, mapped Unity 4.x–5.6 build internals, and reconstructed the relevant logic for auditing and rebuilding.
TECHNIQUES
- →Smali / DEX decompilation and analysis
- →Unity 4.x–5.6 build internals (Mono/IL2CPP)
- →Native code inspection
- →Behavior reconstruction

Web Conferencing System
A web-based video conferencing system built on Node.js, MongoDB, and WebRTC — the browser counterpart to the MeetUp desktop client, delivering real-time multi-party audio and video.
CHALLENGE
Delivering low-latency, multi-party audio and video in the browser required managing WebRTC peer connections, signaling, and media negotiation reliably across networks and devices.
SOLUTION
Built a Node.js signaling server with Socket.io, orchestrated WebRTC peer connections for multi-party rooms, and used MongoDB for room and session state.
TECHNIQUES
- →WebRTC peer-connection orchestration
- →Node.js + Socket.io signaling server
- →Multi-party room management
- →Media negotiation across networks

Sales Management System
A product sales management platform built for company operations at NovaCircuit — covering orders, inventory, and reporting for internal sales workflows.
CHALLENGE
Internal sales operations needed a single system tying together orders, inventory, and reporting, reliable enough to run day-to-day business.
SOLUTION
Built a full sales management platform on ASP.NET/C# with SQL Server, covering order entry, inventory tracking, and management reporting for internal teams.
TECHNIQUES
- →ASP.NET / C# business application
- →SQL Server data modeling
- →Order and inventory workflows
- →Management reporting

Internal Management System
A remote internal security and operations management system built at NovaCircuit, giving administrators centralized oversight of internal resources and activity.
CHALLENGE
The company needed centralized, remote oversight of internal systems and security-relevant activity, accessible to administrators without exposing sensitive infrastructure.
SOLUTION
Built internal management and security-oversight tooling on ASP.NET/IIS with Node.js services, providing administrators a centralized dashboard over internal resources and activity.
TECHNIQUES
- →ASP.NET / IIS internal tooling
- →Node.js supporting services
- →Centralized admin dashboards
- →Remote security oversight