Available for software engineering roles - backend systems, production reliability, cloud deployment, and AI-assisted workflows.

Connect
AI Workflow Automationprototype

BackPocket OS AI Offline

Local-first AI operations assistant prototype for small-business admin workflows, turning emails, quote requests, documents, voice notes, and follow-ups into structured tasks with operator approval.

Case file

Main outcome

Defined a local-first product direction for small-business admin automation.

Secondary signal

Modeled messy business inputs as auditable, structured workflows.

Evidence count

3 proof points

01

Problem

Small operators lose time in admin work: quotes, invoices, Gmail follow-ups, job notes, payments, and paperwork. Existing tools can be expensive, cloud-first, keyboard-centric, and disconnected from field work.

02

Solution

Designed the offline direction around an operator-owned business OS where AI can draft, classify, and organize work while the operator approves before anything leaves the system.

03

Role

Prototype product engineer · Workflow automation designer · AI-assisted operations system designer

Impact

  • Defined a local-first product direction for small-business admin automation.
  • Modeled messy business inputs as auditable, structured workflows.
  • Kept privacy, local ownership, and human approval central to the design.

Technical Highlights

  • Email triage pipeline with Gmail/IMAP input, AI classification, pending approvals, and send/archive planning
  • Human-in-the-loop rule for outbound email, quote, post, payment, and destructive actions
  • Local-first storage direction with SQLite first and PostgreSQL/self-hosted path later
  • AI fallback strategy across Ollama, OpenRouter, and Gemini
  • Construction workflow modeling for leads, quotes, invoices, payments, job files, ABN/GST validation, and pipeline visibility
  • Voice command interface direction with screen context and confirmation cards
  • RAG and document-intelligence direction using Google Drive sync and semantic search

Proof

  • Portfolio positioning and repo-audit source of truth
  • Public GitHub repositories for online and offline prototype versions
  • Explicit prototype status and limitations

Constraints

  • Not production SaaS yet; it needs verified real workflow usage before being described that way.
  • External actions need explicit operator approval and clear auditability.
  • Business data privacy and integration permissions must stay central to the design.

Limitations

  • Prototype / experiment, not a verified production operations platform.
  • Needs clearer demo evidence, screenshots, and integration walkthroughs before being treated as flagship production work.
  • Automation quality depends on source data quality, permission boundaries, and human review.

Roadmap

  • Add a walkthrough showing messy email or document input becoming a structured task.
  • Add screenshots or a recorded demo of the operator review flow.
  • Document privacy boundaries, supported integrations, and non-goals.
All Case StudiesProof-of-Work LabGitHubOffline repo

Next conversation

Let's make the next system less fragile.

Open to software engineering roles across full-stack systems, platform and reliability work, workflow automation, and applied AI. I value teams where I can keep learning while contributing to real systems and clear delivery outcomes.

Also open to freelance or contract work across full-stack builds, practical automation, technical SEO, and cloud delivery.

githubgithublinkedinlinkedintwittertwitteremailemail
Get in Touch