Andy Mackenzie

Data platform, BI, and internal-tools engineer architecting analytics systems and AI workflow automation.

15+ years building and scaling data infrastructure, revenue-operations integrations, and intelligent systems. I specialize in turning complex, siloed data into actionable intelligence and automated workflows using Python, Snowflake, and LLM agent orchestration.

Get in Touch View Projects

Professional Experience

Platforms and data systems architected and maintained throughout my career.

Appcues, Inc.

Business Intelligence Architect

Aug 2018 – Present

Sole owner of Appcues' business intelligence and data platform end-to-end: data warehouse, ETL pipelines, modeling, visualization, and the revenue-systems integrations that sit on top of it.

  • Saved tens to hundreds of thousands of dollars annually via ETL performance optimization and warehouse-sizing strategies, reducing Snowflake credit spend by 40%.
  • Reduced daily metric materialization times on 10B+ row tables by 95% using HyperLogLog sketches.
  • Built and deployed an ETL job MCP using Flask, Pandas, Threading, and Sockets managing 100+ integrations across Netsuite, Snowflake, Hubspot, and Salesforce.
  • Built a LangGraph service in AWS Bedrock Agentcore that parses client order forms and generates Stripe subscriptions autonomously.
  • Maintained an AWS Glue pipeline processing terabyte-scale, 100-billion row JSON datasets into Apache Iceberg tables using pyspark.
  • Sole builder and administrator of Appcues' 300+ model dbt environment and Looker ecosystem.

Dell EMC Corporation

Sr. Analytics Developer — VMAX Pricing & Analytics

Mar 2014 – Aug 2018
  • Managed a 3-person analytics team; mentored junior analysts in programming and data science.
  • Reduced reporting time-to-results by up to 95% by building a comprehensive ETL and reporting stack (PostgreSQL, MSSQL, Python, VBA) surfacing through Tableau Server.
  • Automated 50% of the order-fulfillment process via a Zapier-based orchestration system.
  • Built predictive models for install-base lifecycle and opportunity forecasting.
  • Administered a VMware cluster across 4 Dell rack servers and 20+ VMs for internal tooling.

EMC Corporation

Product Manager / Data Analyst

May 2010 – Mar 2014
  • Authored configuration rules and built the tool used for every VMAX Cloud Edition BOM.
  • Modeled thousands of configuration permutations to identify margin impacts and inform pricing.
  • Owned the quarterly business overview deck, 5-year plans, and forecasting.
  • Gathered and documented requirements for automated ordering applications and presales sizing tools.

Personal Projects

Open-source tools, agentic systems, and infrastructure experiments.

GitHub contribution chart for amack87

All My Agents

Native macOS app for managing AI coding agent tmux sessions — session sidebar, embedded terminals, speedrun mode, zoom controls.

Swift

All My Agents Web

Web interface for managing AI coding agent sessions across machines — mobile-first, multi-machine mesh via Tailscale, speedrun mode.

JavaScript Tailscale

Tmux Agent Hibernator

Automatically hibernate and restore idle AI agent sessions in tmux (Claude Code, Codex, Cursor) to save compute.

Python

Looker MCP

Model Context Protocol (MCP) server that builds entire Looker dashboards from natural language prompts.

TypeScript MCP

Technical Arsenal

The tools, languages, and platforms I use to build data systems.

Data & Analytics

Snowflake SQL dbt / dbt Cloud Looker & LookML Python (Pandas, NumPy) Tableau HyperLogLog ETL Design

AI & Agent Tooling

Claude Code OpenAI Codex MCP Server Dev Anthropic SDK AWS Bedrock AgentCore LangGraph Prompt Engineering

Languages

Python JavaScript / TypeScript Node.js SuiteScript SOQL / Apex Bash R

Infrastructure

AWS (Glue, Iceberg, EC2, S3) Docker Tailscale Linux/Unix Admin Salesforce NetSuite Stripe APIs

How I Operate

I build data platforms and internal tools that turn complex, siloed operational data into automated systems. My approach is rooted in understanding how work actually gets done across revenue, finance, and engineering teams.

In my most recent role, this meant acting as the sole owner of the BI and data platform—running everything from the Snowflake/dbt warehouse to the integration layer across NetSuite, Stripe, and Salesforce. It means finding leverage points, like reducing metric materialization times by 95% using HyperLogLog, or building LLM agents that autonomously process client order forms.

I hold an M.S. in Predictive Analytics from Northwestern University, and I'm actively exploring the intersection of modern data infrastructure and autonomous LLM orchestration to push the boundaries of what internal tooling can achieve.

End-to-End Ownership

From raw API extraction to board-level reporting and financial system orchestration.

High Leverage

Optimizing warehouse costs by 40% and replacing manual labor with deterministic code.

Agentic Systems

Pioneering practical applications of Model Context Protocol and autonomous LLM workflows.

System Architecture

Bridging the gap between software engineering, data warehousing, and RevOps.

Ready to Build What's Next?

Whether you're looking to scale your data infrastructure, optimize warehouse spend, or explore autonomous agent implementations, let's talk.

Email Me Connect on LinkedIn