Products

NoCode Workflow

Architected a graph-based multi-agent orchestration framework, replaced 4 weeks of training with minutes of drag-and-drop on a canvas to build customized LLM workflows with traceability, and benefited 1200+ users monthly; listed as a trending space on Hugging Face.

• Reduced processing time by 86% through model routing and by dispatching concurrent LLM batches through Redis-backed Celery workers.
• Solved bugs reported by users and balanced requests against available resources to keep the product meeting real community needs.
• FastAPI + Celery + Redis backend; React/TypeScript (Vite) frontend, deployed to Render.com and HuggingFace Spaces.
• Built-in cost estimator resolves output-token pricing from a Redis cache of the OpenRouter catalogue.
• 500+ stars on GitHub.
• 70+ stars on Hugging Face.

Open NoCode Workflow →

Live on HuggingFace Spaces →

Wage Explorer

Built a Graph and Hierarchical RAG system to retrieve auditable occupation records, and compiled multi-source documents using LLMs to narrow job descriptions down to their actual duties.

• Reduced the error rate by 85% compared to manual lookups in benchmark evaluation, benefiting employers by eliminating 40+ attorney hours and reducing RFE-refiling risks from generic template filling.
• 500+ stars on GitHub.

Open Wage Explorer →

Price Monitor

Lowered the barrier to quantitative trading for non-technical users by architecting an agentic trading framework with customizable agents roles that let agents safely and autonomously fetch data, build context, make and review decisions with human oversight, and cooperate with each other.

• Developed automated trading software using Celery and Redis to cache and process large-scale, multi-source data feeds, supporting 200+ quantitative trading signals generated by statistical models, and deployed on Kubernetes to scale to 1000+ users.
• Coached 500+ non-technical users by demoing platform mechanics and risk analysis during high-volatility periods, presented the product to prospective investors, and translated feedback into product iterations.
• Extraction agents pull market data, social media signals, and macro inputs; analysis agents synthesize context; execution agents make trading decisions or send notifications.
• Orchestration layers are configurable, so the same framework supports different data sources, decision strategies, and output channels without rebuilding.
• 100+ stars on GitHub.

Open Price Monitor →

Multi-modal AI Chat Platform

A consumer AI platform for chatting with AI characters, generating images and videos, and creating original content, powered by multi-source generative AI APIs across image generation, video generation, and conversational AI.

• Scaled to hundreds of millions of user interactions as co-founder and primary engineer.
• Built the full stack end to end: character systems, a UGC layer, payment and monetization flows, mobile app, and high-concurrency load balancing infrastructure.

AI Tools Directory

A curated directory of 300+ AI tools across 20+ categories including writing, image and video generation, coding assistants, agent frameworks, design tools, and more, with structured comparisons and deployment guides.

• Authored prompt engineering guides, LLM platform comparisons, and developer onboarding tutorials covering deployment, integration, and evaluation.
• Built for developers and startups navigating a rapidly expanding AI tool landscape with structured comparisons and deployment guides for identifying and integrating the right tools.

Distributed VPN Network

A consumer VPN product shipped from zero to production within a small founding team, built and operated alongside two other live products under the same startup.

• Architected a distributed VPN network, utilizing load balancing and multi-region routing across VPS nodes, reducing latency by 90%.
• Co-built with a cross-functional team spanning product, engineering, and operations.