WMS AI Copilot
More info: /projects/wms-ai-assistant
Live URL: https://wms-ai-assistant-web.vercel.app/en
Full-Stack & AI Engineer
🛠️ Tech Stack
- Groq Cloud (Llama 3.x)
- Vercel AI SDK
- TypeScript
- Next.js (App Router)
- Express
- Supabase (PostgreSQL)
- Tailwind CSS
- shadcn/ui
- TanStack Query & Table
- next-intl
- Jest
- npm Workspaces
📖 Overview
Architected and built a full-stack, production-shaped monorepo for a real-time AI copilot designed for modern warehouse management. The project was built for my portfolio instead of a formal job position. The platform enables warehouse admins to perform natural-language queries for stock levels, top sellers, and reorder alerts, returning token-by-token streaming answers or sortable data grids. Engineered a live operational dashboard alongside a tool-calling AI agent, prioritizing data grounding, tight execution boundaries, and seamless internationalization.
🚀 Key Achievements
- Engineered Real-Time Streaming & Tool-Calling Agent: Integrated Vercel AI SDK and Groq Cloud (Llama 3.x) to deliver low-latency, token-by-token streaming responses with deterministic database tool selection.
- Designed Adaptive Visualization Pipeline: Developed an intelligent frontend engine that dynamically renders plain text, or TanStack sortable tables based on the schema of the query result.
- Built Full Operational Split-View Dashboard: Designed a responsive Next.js App Router workspace featuring a resizable split view for real-time inventory, product, and order history tracking.
- Established Security & Trust Boundaries: Enforced a secure LLM access model by restricting the AI to curated, strictly typed DB access tools rather than raw query execution.
- Implemented End-to-End Internationalization: Architected the application from the ground up with next-intl to support fully bilingual (English/Spanish) conversational and UI flows.
- Maintained High Monorepo Code Quality: Structured the project as an npm workspaces monorepo with strict TypeScript typing across API, frontend, and AI tool definitions, backed by Jest test coverage.
We Are Pipol (MIA)
Senior Frontend Developer
🛠️ Tech Stack
- React
- TypeScript
- Next.js
- Tailwind CSS
- Apollo Client (GraphQL)
- AI / LLM APIs
- Mermaid.js
- D3.js
- Virtualized Lists & Tables
- Markdown Rendering
- REST APIs
📖 Overview
Developed AI-powered web applications focused on productivity, conversational experiences, and data visualization. Built modern interfaces that streamed AI responses in real time while rendering rich content—including diagrams, tables, code blocks, and large datasets—with a strong emphasis on performance, usability, and responsive design. Worked closely with product managers and UX designers to rapidly prototype and iterate on AI-first experiences.
🚀 Key Achievements
- Developed a ChatGPT-style conversational interface with real-time streamed AI responses, enabling incremental rendering for a fluid user experience.
- Built a robust markdown rendering pipeline supporting headings, code blocks, syntax highlighting, lists, callouts, tables, and rich formatting.
- Implemented Mermaid.js integration, allowing AI-generated flowcharts, sequence diagrams, ERDs, Gantt charts, and architecture diagrams to render directly inside conversations.
- Designed a flexible renderer capable of displaying AI-generated interactive tables, reports, and structured documentation.
- Implemented virtualized tables capable of rendering tens of thousands of rows with minimal memory usage and smooth scrolling.
- Optimized frontend performance through memoization, virtualization, lazy rendering, and efficient React state management.
- Developed reusable AI chat components supporting streaming, conversation history, message regeneration, loading states, citations, and error recovery.
- Collaborated closely with UX designers to build polished, responsive, and accessible interfaces for AI-powered workflows.
- Leveraged AI development tools—including Cursor, GitHub Copilot, Claude, and ChatGPT—to accelerate implementation, prototype features, and improve code quality.
⭐ Highlights
🤖 AI Chat Experience
- Real-time streamed responses
- Markdown rendering
- Syntax-highlighted code blocks
- Message regeneration
- Loading indicators
- Conversation history
📊 Mermaid Diagram Rendering
AI-generated diagrams rendered directly inside the chat, including:
- Flowcharts
- Sequence Diagrams
- Entity Relationship Diagrams
- Gantt Charts
- Architecture Diagrams
📈 Large Dataset Visualization
High-performance rendering of AI-generated tabular data using virtualization. Features included:
- Sticky headers
- Column resizing
- Sorting
- Filtering
- Infinite scrolling
- Rendering tens of thousands of rows without UI lag
🎨 Rich AI Responses
Support for rendering:
- Markdown
- Tables
- Code snippets
- Mermaid diagrams
- Lists
- Callouts
- Interactive documentation
🔗 Links
🌐 Get MIA: https://get.mia.tech/
Errepar
Frontend Developer Lead
🛠️ Tech Stack
- React
- Next.js
- TypeScript
- Material UI
- Storybook
- Sass
- REST APIs
- Server-Side Rendering (SSR)
📖 Overview
Developed and maintained a large-scale enterprise platform used daily by accounting, tax, legal, and HR professionals across Argentina. Built high-performance applications with React and Next.js, delivering thousands of dynamically generated news articles, legal documents, and professional publications. Focused on scalability, SEO, accessibility, and creating polished user experiences while collaborating closely with designers and backend engineers.
🚀 Key Achievements
- Implemented Server-Side Rendering (SSR) with Next.js, improving SEO and page performance by 100% for dynamic content.
- Optimized rendering and API data-fetching strategies, increasing dynamic page performance by 83%.
- Designed and implemented a reusable Storybook component library, reducing frontend development time by 50%.
- Built scalable UI components aligned with a shared Design System, improving consistency across multiple products.
- Developed responsive and accessible interfaces for thousands of dynamically generated documents and articles.
- Led frontend technical initiatives focused on performance, scalability, and long-term maintainability.
- Established frontend quality standards through code reviews, best practices, and cross-functional collaboration.
- Leveraged AI-assisted development tools to accelerate development and improve engineering productivity.
🔗 Links
🌐 Errepar Site: https://errepar.com/
AI-assisted development
Besides my AI-related projects, I spent the last 2 years working everyday with AI agents (Claude Code, GPT, Cursor, Antigravity) scaffolding solutions in no time then iterating over those solutions by receiving feedback and working over the feedback. Developing quality solutions faster.