Skip to content
  • LinkedIn
  • Github
  • YouTube
Jason Burns
  • About
  • Projects
    • Sumline
    • Cruxwire
    • Backyard Birds
    • Modular Riffs
    • Ceres Garden Assistant
  • Contact

TimescaleDB

  • Home
  • TimescaleDB
The LLM Tax, Revisited: Cost, Latency, and Where to Run the Model

The LLM Tax, Revisited: Cost, Latency, and Where to Run the Model

Quality, cost, and latency. You can usually have two. Here is the framework I use to pick which two on purpose, with a voice assistant, an RSS scorer, and a Frigate box as the worked examples.
August 12, 2026Posted inAI Projects, Product Thinking
Ceres Update: From Empty Scaffold to a Garden That Talks Back

Ceres Update: From Empty Scaffold to a Garden That Talks Back

Seventy-four commits across one weekend took Ceres, the multi-agent assistant for Kim's garden, from a bare scaffold to a working assistant that transcribes voice notes, diagnoses plant photos, answers questions, plans the watering, and runs a scheduled daily briefing from an orchestrated roster of specialist agents.
June 15, 2026Posted inAI Projects, Development, Home Automation, Homestead
Backyard Birds: Building My Own Dashboard on Top of Local Bird Data

Backyard Birds: Building My Own Dashboard on Top of Local Bird Data

A few weeks back I wrote about replacing my Haikubox with a four-camera local pipeline feeding BirdNET into TimescaleDB. The data was beautiful. The dashboards were not. Grafana is great for whether the heat pump cycled overnight or how warm…
May 24, 2026Posted inBirding, Development
Why I Swapped InfluxDB for TimescaleDB – The LLM Tax

Why I Swapped InfluxDB for TimescaleDB – The LLM Tax

For years, InfluxDB has been the data backbone of my Home Assistant setup. With the native Home Assistant integration and Grafana, I had my dashboards dialed in. While FluxQL was a bit of a learning curve, I was able to…
April 24, 2026Posted inAI Projects, Home Automation
PM by trade. Builder by nature.

 

Product manager. Microsoft, Tableau, Elastic. I’m on sabbatical, using the time to build AI systems end to end instead of reading about them: agent orchestration, vision, speech, tool use, evaluation harnesses, local and frontier models working together. Each project here comes with a deep dive into what I built, how it works, and why I made the calls I made.

  • AI Projects 25
  • Birding 2
  • Carpentry 10
  • Computers 8
  • Development 16
  • Home Automation 33
  • Homestead 18
  • Music 2
  • Photography 7
  • Pond 3
  • Product Management 7
  • Product Thinking 6
  • Productivity 5
  • Python 8
  • Rainier 2
  • Security 7
  • Synology 2
  • Updates 5
Tags
Agentic AI AI Product Career Ceres Chicken Coop Cisco Claude Claude Code CoffeeTinker Coral TPU Design Docker Dozzle ECOWITT Frigate Gitea Home Assistant Homelab HomeNetwork LLM Local AI Machines MCP MQTT Music MySQL NAS NetData Network nVidia Obsidian Notes Ollama Open WebUI photography Plex Pond Python PythonNerdery Rainier Raspberry Pi SQL Synology TimescaleDB Ubuntu Voice Assistant
Copyright © 2026 — Jason Burns | Jason Burns
Scroll to Top