Project 01 Work in progress
Nightjar
A work-in-progress case workspace for reading the record, thinking with machines, and keeping hold of the useful work.
- Private for now
- Sources stay close
- Still changing
Neil Williams · Cleveland, Ohio
Hi, I’m Neil.
I work in legal technology, make projects with AI tools, and write about what I’m learning along the way. This is the place where all of that lives.
Right now
July 2026- 01 Nightjar — something new for working a case
- 02 GraftonHub — useful local information, kept current
- 03 DishCache — a better home for saved recipes
Projects
Things I’m working on.
Project 02 Live community project
GraftonHub
A free, no-tracking community information hub for Grafton and southern Lorain County, built around useful local answers and reliable sources.
- Weather + alerts
- Community guides
- No ads or tracking
Project 03 Live personal project
DishCache
A personal recipe vault for saving, cleaning up, searching, cooking from, sharing, and exporting the recipes worth keeping.
- Link import
- Cooking-first views
- Portable recipe library
Writing
Project notes, experiments, and things worth explaining.
The writing is part of the work: what I tried, what changed, and what I learned from making it real.
Building an internal AI litigation workbench
A local-first AI workbench for litigation teams, organized around files, progress, work products, source context, and human review instead of model choice and prompt tricks.
Read article
DishCache: Saving the Recipe, Not the Life Story
DishCache is a recipe app built around a simple idea: save recipes you care about, capture the useful parts from links, and strip away the bloated blog narrative that gets in the way.
Read article
Making a discovery production queryable with local retrieval
Notes from building a local pipeline for a Relativity-style export: parse the load file, normalize the text, index useful chunks, and make the model answer from retrieved documents.
Read article
What I do
Legal technology, useful systems, and hands-on problem solving.
Legal technology & litigation support
Years spent making discovery, trial, infrastructure, and day-to-day legal technology work better for the people using it.
Projects with AI tools
Turning specific ideas into working projects by pairing domain knowledge and product direction with AI engineering tools.
Local AI & automation
Practical experiments shaped around confidential workflows, local control, useful outputs, and human review.
Media, discovery & evidence workflows
Work involving documents, transcripts, audio, video, load files, and the messy handoffs around evidence.
Current work
What has my attention right now.
Finding the shape of Nightjar: a calmer, stranger way to work a case with AI close by.
Keeping GraftonHub current, trustworthy, and genuinely useful to nearby communities.
Improving DishCache as a clean personal recipe library that respects its users.
Writing down what I learn while making and operating these projects.
How I make things
Domain experience in front. AI engineering underneath.
I’m not a software engineer. I know the work, spot useful problems, direct the tools, and keep testing until the idea becomes something real.
- 01Find the gapStart with a problem worth fixing.
- 02Direct the toolsUse AI as the engineering layer.
- 03Work the resultTest, question, and refine it.