Builds

Things that run.

A short shelf of things I've built to answer questions I couldn't leave alone. Start with TinyDesk for workflow design, human review, and evaluation. For workplace agents and employee coaching, see AI in practice →.

TinyDesk ticket view for fictional ticket QE-01, “VPN works in the office and fails at home,” from requester Nadia Pell. On the left: the description, editable Category, Priority, and Status menus with a Save changes button, three activity entries, and a reply composer holding the suggested draft with its own Save reply button. On the right: the navy TinyTalk Assist panel labeled Cloud xAI Grok, with an issue summary, suggested category Network, one question about whether other internet works at home, a draft response, and Apply category and Use draft buttons. Applied AI · Help desk

TinyDesk

A small help desk for working a queue of fictional Ashmere Press tickets. TinyTalk Assist summarizes a ticket, suggests a category, flags what’s still missing, and drafts a reply you can edit. Nothing saves until you click save.

JavaScript front end, Python backend. Local Ollama by default, cloud xAI Grok if you opt in. In a small 12-ticket eval scored for factual errors, repeated questions, and unsupported promises, low-reasoning Grok gave 10 usable drafts and 2 that needed edits.

Screenshot: real interface, fictional ticket. The assistant panel replays a recorded Grok result, not a live call.

Read the case study →

JavaScript · Python · Ollama · xAI Grok
Local AI · Memory

TinyTalk

A small local chatbot that became a practical test of memory reliability. Explicit facts, fact history, a current profile, and source status stay separate from recent conversation so “what is current,” “what used to be true,” and “what was supplied” do not collapse into one transcript.

Read the case study →

Ollama · Python · MemPalace · GitHub ↗
Agent systems

Cricket Mode

A portable behavior layer for AI coding agents. Clamp scope, challenge assumptions, prove the work with evidence, then clean up the residue — across Codex, Cursor, and Antigravity.

Read the case study →

Portable v1 · GitHub ↗
Commit City hero still: a dark skyline of green buildings at different heights. The project README captions this image as the cyb3rcricket contribution skyline. Creative technology

Commit City

Turn a year of GitHub contributions into a 3D city. Every day becomes a city lot, activity becomes height, and the original contribution history stays underneath the spectacle.

Read the case study →

Three.js · TypeScript · Live demo ↗ · GitHub ↗
Mutiny’s real console with a fictional workshop note and test responses, showing its closed research worksheet and Memory panel. Local AI · Privacy

Mutiny

A localhost AI console with Ollama chat, SQLite threads, saved facts, and local jobs. Research is an explicit action, with local-note and web modes and source records attached to each run.

Read the case study →

Ollama · FastAPI · SQLite · GitHub ↗
Grounded AI

LoreMasterBot

A terminal World of Warcraft companion with tools for Blizzard’s official structured data. Local or cloud models share the same lookup loop, with tests for routing, tool-call history, and API failures.

Read the case study →

Python · Tool calling · Blizzard API · GitHub ↗
Mutation Microscope’s real sequence inspector and molecular-impact panel, displaying the bundled APOA1 example. Applied AI · Data visualization

Mutation Microscope

A genomic visualization workspace with chromosome, region, and sequence views, REF/ALT comparisons, and inspectable provenance. Seven curated AlphaGenome examples distinguish published values, reconstructed tracks, and illustrative controls.

Read the case study →

React · TypeScript · AlphaGenome · Live demo ↗ · GitHub ↗