82W DigitalStart a conversation

— 82W Digital

Shipping AI into
places it usually
doesn't go.

Regulated-industry prototypes. Databricks AI agents and apps. Unity Catalog enrichment. Value realization. Senior hands on every engagement, end to end.

Start a conversationSelected work →

What we do

We work where AI has to be rigorous, auditable, and shipped. Regulated industries. Behavioral systems. Data platforms that carry real weight.

A slide deck won't save you. A demo that works on the sample data won't either. The problems we take on are the ones where the system has to hold up the first time, in the room where decisions get made, with the people who have to live with the answer.

Senior attention from first call to handoff. No junior swap-outs, no staff augmentation, no subcontracted work.

Practice areas

Two practices. One way of working.

Everything we take on sits in one of two lanes. Both lean AI-first, both ship working code, and both are run end to end by the same senior hands.

Regulated, rigorous, shipped.

AI Prototyping Studio

We build AI systems for places where it has to work the first time — clinical settings, behavioral systems, regulated workflows. The kind of problems where a demo isn't enough and a slide deck won't save you.

  • Regulated-industry AI

    HIPAA-compliant builds, auditable pipelines, PHI-aware architectures. We've shipped into behavioral health and education environments where the rules are real.

  • Constitutional AI with human oversight

    Rules-based guardrails, escalation paths, and review loops that keep humans in the decision — without making the system useless.

  • AI agent development

    Task-shaped agents, tool-use systems, and multi-step workflows. Built to be inspected, not just invoked.

  • Rapid prototyping to production

    Working code in weeks, not quarters. We ship into real hands early so you learn before you spend.

Value realization for your lakehouse.

Databricks Practice

Most Databricks investments are underused. We help teams close the gap between the platform they bought and the outcomes they expected — with tooling, enrichment, and applications that make the data useful to the people who need it.

  • Unity Catalog metadata enrichment

    Governed metadata at scale. Lineage, classification, PII/PHI tagging, and semantic enrichment that makes Unity Catalog actually searchable — not just populated.

  • Value realization

    Outcome measurement frameworks that prove what the lakehouse is worth. We instrument the workflows, quantify the lift, and hand leadership something they can act on.

  • AI agent development on Databricks

    Agents built on Mosaic AI Agent Framework and Genie. Tooled into your data, governed through Unity Catalog, deployed into the places your team already works.

  • Databricks Apps

    Custom applications on the Databricks Apps platform. Internal tools, decision surfaces, and embedded experiences that live next to the data instead of in a separate stack.

Selected work

Things we've actually shipped.

Four projects that prove the pitch. Open-source, regulated, responsible, and architectural — one each.

01 · Shipping product

NextDialog

A calm interface for AI coding agents.

Multi-session desktop terminal manager for Claude Code, Cursor Agent, Gemini CLI, and custom agents. Built, shipped, and live at nextdialog.io.

Version
v0.3.2
Releases
17
License
MIT
Platforms
macOS · Linux · Windows
02 · Regulated-industry fluency

CareHub

HIPAA-compliant behavioral health SaaS.

A multi-tenant platform for behavioral health providers — clinical workflows, risk flagging, and audit trails that meet the rules without getting in the way of care.

Compliance
HIPAA
Domain
Behavioral Health
Scope
Clinical + Admin
03 · Responsible AI at real stakes

AskLisa AI

Constitutional AI with human oversight for Digital Citizen Academy.

An AI tutoring and guidance system built for an education partner — with rules, escalation paths, and human review loops that keep the humans in charge.

Model
Constitutional AI
Oversight
Human-in-loop
Sector
Education
04 · Graph knowledge architectures

Substrate

A personal cognitive data substrate.

A graph-based knowledge management system with spreading activation, designed to hold and connect personal and project context across time.

Architecture
Graph-native
Algorithm
Spreading activation
Scope
Cognitive infrastructure

How we work

Three phases. No surprises.

Every engagement runs through the same shape. Fixed fees where it matters, capped time-and-materials where it helps, and a written decision point between every phase.

  1. 01

    Discovery

    2–3 weeks

    We map the problem, the constraints, the stakeholders, and the first thing worth building. You leave with a written brief, an architecture, and a decision: move forward, or don't. Fixed fee. No pressure.

  2. 02

    Prototype

    4–6 weeks

    We ship a working version into real hands as quickly as the problem allows. Instrumented. Reviewed. Iterated. By the end, you have something you can show a board, a regulator, or a skeptical engineer — and it works.

  3. 03

    Handoff

    Capped T&M

    We leave you with working code, documentation that matches the code, and a team that can run it without us. Optional production hardening and follow-on engagements, always scoped and capped.

Fit

Who this is (and isn't) for.

◆ Good fit

  • Teams where AI has to be rigorous, auditable, and deployable the first time.
  • Healthcare, behavioral health, education, and other regulated or high-stakes domains.
  • Data teams trying to extract real value from an existing Databricks investment.
  • Founders who need a working prototype in weeks, not a decision framework in months.

◇ Not a fit

  • Staff augmentation or body-shop engagements.
  • AI readiness assessments, maturity models, or 20-page procurement roadmaps.
  • Projects that need a sales team, a marketing team, or a subcontracting layer.
  • Teams that haven't decided whether AI is worth doing. That's a different conversation.

— Start a conversation

Got a problem worth
shipping an answer to?

One email. Describe the problem, the stakes, and the timeline. We reply to everything that isn't a sales pitch.

Start a conversationbrian@82wdigital.com