Selected work for applied AI and platform engineering roles
Selected work ordered for hiring decision makers: applied AI operations systems, customer-facing app-builder software, trusted knowledge infrastructure, workflow SaaS, founder/business credibility, live product delivery, and regulated engineering.
Multi-agent operations layer
My OpenClaw company system
Built an internal multi-agent operations control plane for running real business workflows with specialist agents, durable memory, scheduled workflows, browser and messaging automation, command controls, metrics, and guarded execution.
Designed a coordinator and specialist-agent operating model with persistent workspace state and clear ownership boundaries
Built command-console controls, cron workflows, browser automation, messaging workflows, and operational memory
Applied the system to product work, release discipline, research, monetization follow-through, and attention allocation
Built applied-AI operations infrastructure around real business workflows, approvals, memory, tools, and execution loops
AI app launch cockpit
Built a customer-facing AI app planner, safe private-preview builder, and ownership cockpit for businesses that want real web apps they own.
Built an AI App Planner that turns plain-English app ideas into reviewed launch plans and scoped private-preview requests
Designed a preview-first build path with isolated branches/checkouts, structured AI patching, tests/builds, Vercel previews, smoke checks, readiness summaries, and restore points
Built app ownership cockpit surfaces for connected accounts, provider resources, launch checklist, domains, app health, audit logs, alerts, usage, ownership, and production approval boundaries
Designed tenant/workspace/app isolation, token-vault boundaries, provider-connector rules, and fail-closed production safety posture
Applied AI infrastructure / open-science compute
Built a distributed volunteer-compute platform for open biomedical literature processing, combining a web coordinator, Windows desktop worker, local LLM execution, signed work packets, validation, public impact metrics, and trust-oriented volunteer controls.
Built coordinator APIs for worker registration, signed packet claiming, result submission, release/fail handling, heartbeat, worker control, and admin monitoring
Implemented local desktop worker flows for Ollama/model readiness, resource limits, idle/battery controls, pause/resume, activity timeline, and enrollment/profile setup
Designed validation and evidence handling around structured outputs, citation/provenance retention, format checks, consensus status, and researcher-review posture
Added public trust surfaces: impact metrics, leaderboards, volunteer profiles, responsible disclosure, science disclaimer, download verification, and early-access caveats
Hardened production storage paths, admin surfaces, enrollment flows, ingestion workflows, abuse/reporting surfaces, and public metric freshness
Worker dashboard
Volunteer-facing control and transparency surface for local LLM work.
Desktop worker
Resource controls
Trust-oriented settings that keep contribution bounded by volunteer preferences.
Volunteer control
Trusted AI knowledge infrastructure
Built paired AppAssist products covering the full trusted-knowledge path: Forge turns approved business documents into cited, reviewable knowledge; Bridge delivers approved knowledge as cited production answers through APIs, widgets, chatbots, and internal tools.
Bridge covers knowledge bases, retrieval, release lifecycle, API keys, widgets, answer/chat/retrieve APIs, analytics, evals, RBAC, and hosted launch hardening
Validated Bridge production readiness with migrations, indexing, release publish/rollback, widget origin checks, API retrieval, RBAC negative checks, and CI-backed launch documentation
Built a production-oriented AI knowledge product stack with source provenance, citations, trust controls, APIs, onboarding, billing/pricing posture, and operational validation
Service-business revenue workflow
Built paired workflow products around the service-business revenue lifecycle: Tavrio for controlled missed-lead, proposal, and payment follow-through, and DealOps for proposal approval, payment progression, and kickoff handoff.
Built and led a patented market-research and engagement platform from concept to scale, raised $500K+, and grew to more than 20,000 registered participants.
Founder-level ownership across product, engineering, operations, and platform evolution
Built and operated the platform long enough to prove product persistence, founder ownership, and execution range
Turned ambiguity into a real, scaled software business
RAADZ site
Methodology-led product framing around private preference, market perception, and research discipline.
Founder platform
RAADZ template library
Productized methodology surface showing reusable research patterns, decision support, and a developed platform model.
App workspace
RAADZ survey setup
Configurable survey and research setup workflow inside a working product with reusable research patterns.
App workflow
Live hosted product
Built and operated a live hosted app for golf pool administration and participant-facing workflows, showing practical AppAssist-style product execution in a concrete deployed product.
Built commissioner/admin workflows for hosted pool setup and management
Built participant-facing views and mobile-friendly recap flows
Built and operated a live hosted product with admin workflows, participant UX, mobile views, and operational polish
Golf Pools admin
Commissioner/admin workflow showing a polished operator experience for hosted pool management.
Hosted app admin
Golf Pools site
Clear product positioning for hosted golf pools with a practical owner/admin workflow.
Product site
Golf Pools mobile recap
Mobile-friendly participant view that adds broader product-design range beyond workflow SaaS.
Mobile product view
Regulated product engineering
Tour Trader Pro / Player Options
Built regulated wagering products across Flutter mobile, React web, PHP + SQL Server backend, and Azure infrastructure, including two GLI-33-certified products.
Took Tour Trader Pro from idea to MVP in under two months
Owned real-time pricing, operational workflows, API integrations, and production ops support
Shipped under compliance, reliability, and audit-sensitive constraints in a regulated product environment
Player Options mobile product
Data-dense mobile product design in a regulated sports environment with strong information hierarchy and live-state clarity.
Regulated mobile product
Player Options lineup workflow
Production mobile UX with dark-mode information density, structured contest flows, and polished scanability.
Regulated mobile product
Supporting product portfolio
Additional products and experiments
Additional shipped properties and experiments show range across product packaging, decision-support workflows, sales automation, commercialization surfaces, consumer mobile, and technical R&D.
