Local-first AI systems

I build the machine that builds things.

My portfolio is not a pile of demos. It is a connected AI workbench: private assistants, autonomous coding, memory, verification, reliable tool calls, document intelligence, code intelligence, and reusable agent skills.

See the workflow Craig demo site

15+ connected AI projects

Local privacy-first architecture

Real apps, CLIs, and agents

I started with a simple problem: AI is powerful, but most AI tools are too cloud-dependent, too forgetful, too unreliable, or too abstract to trust with real work.

So I built the pieces I needed internally. HammerLock is the workspace. Craig does the building. The supporting tools handle memory, verification, scraping, prompt compression, code edits, code review, and tool-call reliability. Together, they form a local-first AI operating layer.

The Stack

Each layer exists because production AI agents fail in predictable ways: they forget, hallucinate, call tools badly, overwrite too much code, lose source context, and burn too many tokens.

01 Product

HammerLock

Private AI assistant with encrypted memory, vault storage, and specialist agents.

02 Builder

Craig

Autonomous software engineer that inspects workspaces, edits files, and runs commands.

03 Memory

RecallMax

Compresses long conversations into decisions, facts, errors, and open questions.

04 Trust

Guardian + Verifier

Retries failed tools, repairs JSON, checks URLs, validates outputs, and scores trust.

05 Context

Docs + Repos

Local RAG and codebase intelligence give agents grounded private knowledge.

A Real Workflow

One request can move through the whole system: research, verify, compress, remember, draft, check, and store the result privately.

User:
Research competitors and draft a pitch deck.

HammerLock:
Routing to Strategist agent...

Web Scraper:
Collected competitor pricing, product pages, and public claims.

Real-Time Verifier:
Checked source URLs and structured data.

Prompt Condenser:
Reduced research context while preserving URLs and facts.

RecallMax:
Loaded company positioning and prior decisions.

Tool Use Guardian:
Wrapped calls with retries, timeouts, and JSON repair.

HammerLock:
Draft saved to encrypted vault.

Ask HammerLock for work, not chat.

The user gives a real outcome: research competitors, draft a deck, inspect a repo, or build a feature.

Give the agent private context.

Local docs, repo structure, prior memory, user preferences, and project history stay under user control.

Make every tool call observable.

Retries, timeouts, repair metadata, validation checks, and source trust scores make the workflow less fragile.

Ship the result back into the vault.

The output is saved where the user can reuse it, audit it, and feed it into the next task.

Demo Lab

The site should not only describe the projects. It should show what they do through small, concrete workflows people can immediately understand.

HammerLock

Private Strategy Assistant

A founder asks for competitor research and gets a sourced draft saved into an encrypted local vault.

Task:
Research three competitors and draft a pitch-deck outline.

Agent:
Strategist selected.
Sources collected: 12
Verified URLs: 12/12
Memory loaded: positioning, ICP, prior notes

Output:
1. Market wedge
2. Competitor table
3. Differentiation slide
4. GTM narrative
5. Risks and proof points

Product site Repo

Craig

Autonomous Feature Build

Craig reads a codebase, plans the change, edits files, runs checks, and reports what changed.

User:
Add PDF export to the report page.

Craig:
Scanning app routes...
Found report component.
Editing export action.
Adding download button.
Running build...

Result:
3 files changed
Build passed
Next: verify generated PDF layout

Craig site Repo

Agent Infrastructure

Reliable Tool Chain

Tool Use Guardian, Real-Time Verifier, and Prompt Condenser turn fragile agent steps into observable ones.

Before:
API timeout
Malformed JSON
Dead source link
Prompt too large

After:
Retry 2 succeeded
JSON repaired
11/12 links verified
Context reduced 38%
Failure metadata attached

Guardian Verifier Condenser

Context Layer

Private Docs + Codebase Q&A

Local Doc RAG and Repo Intelligence let agents answer from private PDFs, DOCX files, and source code.

Question:
Where does the app validate license keys?

Repo Intelligence:
lib/license-keys.ts
app/api/license/validate/route.ts
tests/license-keys.test.ts

Local Doc RAG:
Policy source: security-runbook.pdf
Relevant section: offline activation rules

Repo Intelligence Local Doc RAG

Design Range

Each project should feel like it belongs to its user and job: secure vault, coding workbench, marketplace, code map, document research desk, and practical business dashboard.

⌁

Secure dark product

HammerLock AI

Premium privacy assistant: encrypted memory, local vault, and user-controlled workflows.

READroutes, scripts, components

EDIT3 files changed

CHECKbuild passed

Light developer workbench

Craig

Practical coding agent: reads repos before writing code and shows the diff.

Verifier42 FLUX

Scraper18 FLUX

Editor64 FLUX

Futuristic marketplace

Genesis Marketplace

Agent skill economy: browse, price, install, and trade capabilities.

Repo Intelligence

Architecture-first interface for understanding private source code with citations.

Source: policy.pdf, p. 12

Document research desk

Local Doc RAG

Calm knowledge UI for private PDFs, DOCX files, citations, and source excerpts.

SynthPipe CRM

Dense small-business pipeline UI for contacts, follow-ups, and local AI sales help.

Craig the Builder

Websites, CRM, and AI follow-up for local businesses. Craig finds the digital gaps, builds the missing stack, and gives busy owners a simple system they can actually use.

Lead finder

Digital readiness audit

Craig scans a business website or public listing and turns the gaps into a simple action plan.

Mobile siteweak Lead formmissing Local SEOthin CRM follow-upnone

Modern Website

Fast one-page or small-site build with services, trust signals, calls to action, and local SEO structure.

Lead Capture

Contact forms, quote requests, appointment links, missed-lead tracking, and clean intake fields.

Simple CRM

SynthPipe-style pipeline for new leads, follow-ups, booked calls, open estimates, and won customers.

AI Follow-Up

Email templates, review requests, reactivation messages, and next-step suggestions for busy owners.

Dental / medical

Turn searches into booked appointments

A modern services site with appointment CTAs, insurance/payment notes, review prompts, missed-call follow-up, and local SEO pages.

booking CTA review flow patient FAQ

Contractors

Capture quote requests before competitors do

A fast mobile site with service area pages, quote forms, photo proof, job-type routing, and follow-up reminders for open estimates.

quote form service areas estimate CRM

Med spas / salons

Promote offers and rebook clients

A polished offer page, treatment menu, before/after gallery, lead magnet, reactivation emails, and simple campaign tracking.

offer page gallery rebooking

Accountants

Organize seasonal lead flow

A trust-focused site with tax-season intake forms, service pages, document checklists, lead status, and automated next-step emails.

intake form checklists follow-up

Small law offices

Make consultations easier to request

A professional site with practice-area pages, consultation forms, source tracking, intake triage, and careful privacy-aware messaging.

practice pages consult forms lead triage

Cannabis operators

Explain services in a regulated market

A compliance-aware site with service explainers, market-entry forms, gated downloads, CRM tracking, and multilingual content.

compliance copy downloads CRM

$500 starter build

One-page website, contact form, local SEO basics, and a simple CRM pipeline.

$199/mo operations

Hosting, updates, lead tracking, review prompts, and follow-up templates.

$499/mo growth

Content, landing pages, automations, campaign ideas, and monthly improvement audits.

Proof of Work

The strongest version of this portfolio is demo-led: short videos that show the projects solving real tasks, backed by code and clear tradeoffs.

Record first

HammerLock Research Run

Show the assistant taking a private request, collecting context, verifying sources, and saving the result into the vault.

High signal

Craig Builds a Feature

Show a coding task from prompt to diff to build result. This is the clearest proof that the system can create working software.

Infrastructure

Agent Reliability Demo

Show a broken tool call, malformed JSON, or bad URL becoming retryable, repairable, and verifiable through the infrastructure layer.

What this proves

I can build full-stack apps with TypeScript, Next.js, and local AI workflows.
I can turn vague AI ideas into usable products, CLIs, and developer tools.
I understand reliability, verification, privacy, context, and product value.
I can explain what I built clearly enough for users and teams to evaluate it.

What I am looking for

Junior-to-mid full-stack, AI tooling, automation, or product engineering roles.
Teams that value shipping useful software, learning fast, and clear communication.
Contract projects where local AI, document workflows, agent tooling, or automation can save time.
Opportunities where proof of work matters more than a traditional tech background.

Not another chatbot.

This is the shape of the work: local-first AI that can understand private context, use tools, verify outputs, preserve memory, and help build the next thing.

View GitHub Contact