Product Manager | Integrations, AI & Automation

Building products where users, data, and systems connect

I’m a technical Product Manager with a background in integration engineering, building AI-powered products, enterprise integrations, automations, and data-driven workflows from concept through production.

Recent Projects

Selected product, integration, and technical work

Professional project details have been simplified to protect confidential and proprietary information.

MemoryOS

Cognote

Personal Project

Personal knowledge platform built with semantic search, RAG, and the Groq API.

Next.js Supabase Groq API RAG
Teams Integration

Teams Integration

Live

Enterprise Microsoft Teams integration that brought AI workflows into an existing communication tool.

Teams SDK REST APIs JWT Auth Figma
AI Chat Interface

AI Chat Interface

Live

In-product support experience designed to make onboarding and common product questions easier.

HTML/CSS JavaScript WebSocket REST APIs
Chrome Extension

Chrome Extension

Live

Browser extension that simplified adding web content and files to AI knowledge workflows.

JavaScript Chrome APIs REST APIs Figma
Parrot Website

Parrot Website

Freelance

Complete website rebuild with SEO optimization and brand asset creation.

HTML/CSS JavaScript PHP SEO

About

I build at the intersection of product and engineering

I’m a Product Manager based in San Francisco with a background in integration engineering. My work sits at the intersection of product strategy, AI, automation, APIs, and data, especially when complex technical systems need to become useful, reliable products.

At Personal AI, I progressed from hands-on integration engineering into product management while continuing to work directly with technical systems. I own initiatives from discovery and requirements through implementation, testing, launch, and iteration, partnering closely with engineers and customers throughout the process.

My work has included building enterprise API integrations, AI agents and custom workflows, automation pipelines, analytics and monitoring systems, and internal operational tooling.

I’m most effective on problems that require both technical depth and product judgment. I like understanding the user problem, mapping how systems and data need to interact, and working through the details required to ship something reliable. Outside of work, I experiment with emerging AI and automation tools, build personal projects, mountain bike, and golf.

Product & Strategy

  • Product discovery, requirements & prioritization
  • Technical product & integration strategy
  • Customer research & feedback
  • Cross-functional engineering collaboration
  • Agile development & Jira workflows

AI & Automation

  • AI agents & LLM-powered workflows
  • API-driven automation
  • LangChain & agent prototyping
  • Zapier, n8n & Superblocks
  • Custom enterprise workflows

Integrations & Systems

  • REST APIs, OAuth, Webhooks & JSON
  • Enterprise platform integrations
  • Event-driven workflows
  • API documentation & technical troubleshooting
  • Integration monitoring & reliability

Data & Development

  • Python, JavaScript, TypeScript, SQL & Node.js
  • PostgreSQL, MySQL, Elasticsearch & InfluxDB
  • Data ingestion & transformation pipelines
  • Analytics & operational dashboards
  • Docker, Git, Tableau & Power BI

Get in touch

Available for new opportunities

I’m interested in opportunities across technical product management, integrations, AI automation, and business systems, particularly roles where I can combine product ownership with hands-on technical problem solving.

← Back to work

Chrome Extension Design & Development

2024 Personal AI Product + Browser Extension
Chrome Extension Interface

Problem

Adding web content to the platform required users to leave the page they were viewing and move through a separate upload flow. That added friction to a task users often repeated while researching or collecting information.

Approach

I designed and built a browser extension that let users capture the current page or upload a file without leaving their workflow. The experience prefilled useful page information and kept the upload process lightweight.

Result

The extension reduced the number of steps required to add content and gave users a faster way to build their knowledge while browsing.

Extension popup interface Settings panel design
← Back to work

AI Chat Interface & User Onboarding

2025 Personal AI Product + UX
AI Chat Interface

Problem

New users often had basic questions about setup, settings, and where to find key features. Documentation existed, but searching through it during onboarding added friction.

Approach

I built an in-product support experience that let users ask common questions without leaving the application. I designed it to stay out of the way when it was not needed, with collapsible, draggable, full-screen, and accessibility-friendly states.

Result

The experience gave users a more direct way to get help during onboarding and reduced the need to leave the product to search documentation.

Chat interface flow Onboarding tutorial screens
← Back to work

Microsoft Teams Integration

2025 Personal AI Product + Integration
Teams Integration

Problem

An enterprise customer wanted employees to access AI functionality inside Microsoft Teams rather than asking users to adopt another application for a common workflow.

Approach

I helped define and deliver a focused Teams integration that brought the core experience into the customer’s existing communication workflow. I owned product requirements and UX while partnering with engineering on authentication, API integration, testing, and rollout.

Result

The integration reduced context switching and made the product easier to access inside an existing enterprise workflow. It also created a reusable pattern for bringing product capabilities into third-party platforms.

← Back to work

Cognote

2025 Personal Full-Stack + RAG
Cognote System Architecture

Problem

Notes, research, documents, and ideas tend to pile up across different places, making useful information difficult to find later. Keyword search works when you remember the exact wording, but it is less useful when you remember the idea or context instead.

Approach

I built Cognote as a full-stack RAG system for storing and retrieving personal knowledge. Content is stored in Supabase and converted into vector embeddings for semantic search. When a user asks a question, the system retrieves relevant context from their knowledge base and sends it to a language model through the Groq API to generate a response grounded in their stored content.

Result

The working prototype supports semantic search and RAG-based responses across a user's stored knowledge. Users can ask natural-language questions, retrieve relevant memories even when the wording does not match exactly, and generate responses using their own content as context.

← Back to work

Parrot Website Rebuild

2024 Freelance Full-stack + SEO
Parrot Website

Problem

Parrot Communications's existing website had poor search rankings, outdated design, and lacked mobile optimization. The site failed to convert visitors and didn't reflect the company's brand evolution and service offerings.

Insight

Small creative agencies need websites that showcase their work while being discoverable by potential clients. SEO performance and visual design quality directly impact client acquisition in the creative services industry.

Impact

Organic search traffic increased by 340% within 6 months. Page load speed improved by 60%, and mobile usability scores reached 95/100. The client reported 3x more qualified leads and 40% higher conversion rates from website visitors.