I turn business problems into production AI systems.

Intake Form Pro

Conversational and voice intake that produces a usable brief, not a form submission.

The problem

Legal intake arrives as half-finished web forms and voicemails. Someone still has to call the person back, work out what actually happened, and decide whether the matter is worth taking.

What I built

A SaaS intake product that interviews the client conversationally — by chat or by voice — asks the follow-up questions a paralegal would ask, and hands the firm a structured summary with a lead score attached.

Why it matters to you

Shows the full path from an unstructured human conversation to a structured record a business can act on — the same shape as most enterprise AI intake problems.

AI: AI interviewing, voice AI, automatic summarisation, lead scoring
Architecture: Multi-tenant SaaS, workflow engine, REST integrations into firm systems
Governance: Tenant isolation, role-based access, audit trail on every intake record

Omni Forge / Plainfield

AI implemented inside an institution, with the workflows and the rollout that entails.

The problem

Institutions have plenty of AI enthusiasm and almost no path from enthusiasm to something staff will actually use inside their existing processes.

What I built

An institutional AI implementation covering the education workflows themselves, not just a tool drop — deployed across a district rather than piloted with one team.

Why it matters to you

Enterprise buyers ask the same question a district does: who owns it, who is allowed to see what, and what happens on day two. This is the answer with a real deployment behind it.

AI: Workflow-embedded AI, institutional deployment patterns
Architecture: Multi-user deployment across an institution, integrated with existing education workflows
Governance: Institutional access control and role separation

Adam Legal Systems

A connected legal technology ecosystem rather than a set of disconnected tools.

The problem

Legal practices accumulate systems that do not speak to each other, so the same matter gets re-entered and the same question gets answered differently depending on where you look.

What I built

A connected ecosystem of legal technology with a deliberate architecture underneath it — shared records, orchestrated workflow, and integration between the parts rather than around them.

Why it matters to you

This is enterprise architecture work in a domain where getting it wrong is visible immediately — the closest analogue to an enterprise AI solutions architect brief.

AI: AI-assisted legal workflow across a connected system
Architecture: Ecosystem architecture, workflow orchestration, cross-system integration
Governance: Access control and auditability across connected systems

Compliance Assist AI

AI-assisted compliance operations built to survive an audit.

The problem

Compliance teams are asked to move faster without losing the trail. Most AI tooling makes the first part easy and the second part impossible.

What I built

A secure multi-tenant compliance workflow where AI assists the operator and every step stays attributable — role-based access throughout and an auditable record of what happened and who did it.

Why it matters to you

The exact objection an enterprise raises about AI — 'we cannot show a regulator what it did' — answered with an architecture rather than a policy document.

AI: AI-assisted compliance operations, document handling, workflow assistance
Architecture: Secure multi-tenant SaaS, RBAC, workflow engine, audit logging
Governance: Multi-tenancy, role-based access control, full auditability

YRC AI Receptionist

A voice agent that answers the phone and then does something about it.

The problem

Calls get missed, and the ones that get answered end as a note on a pad that never reaches the system that would act on it.

What I built

A voice agent handling front-desk and contact-centre calls, integrated into the business systems behind it so a completed call becomes a record, a task or a booking rather than a transcript.

Why it matters to you

Voice is where AI either integrates or embarrasses you. This is the integrated version, and the pattern generalises to any customer-facing agent brief.

AI: Voice agents, contact-centre automation, intent handling
Architecture: Voice pipeline integrated with business workflow systems
Governance: Call handling with controlled hand-off to business systems

Adam Legal Data / Reporting

The data and reporting foundation the AI work sits on.

The problem

Firms cannot answer basic questions about their own matters, and no amount of AI on top fixes a data layer that was never built.

What I built

Enterprise data and reporting — warehousing, ETL and BI — connected forward into modern AI and analytics rather than left as a separate reporting silo.

Why it matters to you

Answers the question every serious AI buyer eventually asks: what is the data underneath this, and who built it. Twenty years of BI and data work is the differentiator, not a footnote.

AI: Analytics and AI built on a governed data foundation
Architecture: Data warehousing, ETL, SQL, BI and dashboards
Governance: Governed reporting access

AI Discovery Assessment

$3.5K–$7.5K

A ranked list of where AI actually pays inside your business, and what it would take to build it.

Two to three weeks working through your workflows to find the ones worth automating, ranked by return and by how hard they are to build.

• — Workflow analysis
• — Opportunity assessment
• — ROI ranking
• — Architecture recommendation
• — Roadmap

AI Discovery + Roadmap

$7.5K–$15K

A prioritised implementation plan with a business case your board will sign off.

The assessment, extended into a sequenced plan: which use cases in which order, what each costs, what each returns, and the architecture that carries all of them.

• — Prioritised use cases
• — Implementation plan
• — Business case
• — Technical architecture

AI Prototype Sprint

$10K–$20K

A working prototype of your highest-value use case, running against your real workflow.

One use case, specified and built as an integrated working demo — the fastest way to settle an internal argument about whether this will work here.

• — Use-case specification
• — Working prototype
• — AI integration
• — Workflow demonstration
• — Production roadmap

AI Agent Implementation

$15K–$40K

A production agent doing real work inside one of your workflows.

Design and build of a production workflow agent — support, sales, intake, document processing or internal knowledge — integrated with the systems it has to touch.

• — Agent design
• — Systems integration
• — Human review workflow
• — Deployment
• — Handover and support

Custom AI Application

$25K–$100K

A complete AI-enabled application, designed, built and deployed.

Full custom build — SaaS or internal — where an off-the-shelf tool cannot reach the workflow you actually run.

• — Product definition
• — Architecture
• — Full build
• — Security and access model
• — Deployment and operations

Fractional AI Architect

$125–$175 per hour

Senior architecture and implementation leadership on your AI work, without a permanent hire.

Ongoing architecture, technical advisory and implementation leadership for teams building AI who need someone accountable for the design.

• — Architecture ownership
• — Implementation leadership
• — Technical advisory
• — Team review

Fractional AI CTO

$8K–$20K per month

AI and technology strategy owned end to end, at a fraction of a CTO hire.

Strategy, architecture and execution leadership across your AI and technology programme on a monthly engagement.

• — AI and technology strategy
• — Architecture ownership
• — Execution leadership
• — Board and exec reporting