Skip to content
The work

Built, measured, and honestly scoped.

30 projects across the range, from frontier-model research teams to a robot on a USDA line. 24 were built or productionized; the rest were architected, integrated, configured, or led. The tag on each card makes the attribution explicit.

All projects

ProductionizedEnterprise Mobility

Governed Research Department

One request becomes an independently reviewed decision package

Productionized open-source Paperclip as the control plane for a governed research organization: intake, parallel evidence lanes, citation review, independent assessment, adversarial challenge, and human release. In its first internal pilot, six parallel research lanes produced a cited decision package; independent review supported all 16 sampled claims and caught two material errors before release. Clippy provides the guarded intake and status front door without making decisions for the user.

  • 6 parallel research lanes
  • 16/16 sampled claims supported
  • 2 material errors caught
  • human release gate

Paperclip · Hermes · OpenWebUI · Azure AI Foundry · MCP

BuiltEnterprise Mobility

SaaS-Hermes

A private specialist-agent platform with durable memory

Upgraded and consolidated a private Hermes research fleet around six core specialists, durable Honcho memory, curated RAG, voice, and governed tool access. The platform moved to Hermes v0.17, Honcho 3.0.11, and OpenWebUI 0.10.2 while consolidating six gateways and reducing the application topology from 17 to 12. Production checks covered cited research, identity, RAG, and speech paths in 5 to 17 seconds.

  • 6 core specialists
  • 17 to 12 applications
  • 2,070-item backlog drained
  • 5 to 17s production checks

Azure AI Foundry · Container Apps · Azure AI Search · MCP · WebRTC · FastAPI

ArchitectedEnterprise Mobility

Private Enterprise GenAI Platform

The in-tenant control plane every solution plugs into

A fully in-tenant Azure AI platform where no inference or data ever leaves the company tenant. One governed, SSO-gated workspace delivering multi-model chat, agentic RAG, voice AI, and document intelligence, plus the reusable blueprints that turn one-off builds into a repeatable platform.

  • 14 solutions
  • 131 model deployments
  • ~45 MCP tools
  • private, SSO-gated environment

Azure · Private VNet · Entra SSO · Managed Identity · OpenWebUI · LiteLLM

BuiltMeat Bot

Meat Bot

AI-assisted food-production robotics, operating separately from CBJ

Designed, built, and deployed the first Meat Bot production unit at a USDA-inspected co-packer, then developed the platform across eight workstreams: ESP32-S3 firmware, GRBL motion control, OTA, fleet cloud, command-and-control dashboard, hardware, documentation, and web. The business now operates separately from Chicken Butt Jerky LLC; the first unit has operated at the co-packer since April 2026.

  • first Meat Bot unit deployed
  • operating since April 2026
  • 2 provisional patent applications
  • 8 platform workstreams

ESP32-S3 · KiCad · GRBL · OTA firmware · Google Cloud Run · Supabase

meat-bot.com
ProductionizedMeat Bot + Chicken Butt Jerky LLC

AI Operating System for Founder-Led Businesses

Three specialized operating organizations with explicit human gates

Productionized a self-hosted, config-as-code Paperclip control plane across Meat Bot engineering, Meat Bot sales and research, and Chicken Butt Jerky operations. The system coordinates board workflows, provider lanes, capacity, immutable releases, exports, and human approvals. As of August 12, 2026, the three organizations contained 45 active software and service seats; that count describes configured automation capacity, not employees.

  • 3 specialized organizations
  • 45 active software/service seats
  • config-as-code
  • human approval gates

Paperclip · Hermes · Pi · Git · self-hosted control plane

BuiltMeat Bot

Meat Bot Evidence-Led Sales Research

A governed research corpus, deliberately separated from outreach

Built a dedicated Meat Bot research organization that mapped 75 target organizations and 73 named contacts, including 17 personal emails verified through Apollo. The system prepared evidence and CRM-ready records under human-controlled outbound rules; no campaign was launched and no commercial outcome is claimed.

  • 75 target organizations
  • 73 named contacts
  • 17 verified personal emails
  • 0 outreach sent

Paperclip · Apollo · structured research · human outbound gate

BuiltChicken Butt Jerky LLC

Hermes

A self-hosted specialist-agent layer for founder operations

Built a self-hosted set of specialist agents supporting Chicken Butt Jerky across operations, marketing, finance, retail intelligence, internal tools, and quality review. Shared memory and governed integrations let the agents carry context across work while keeping consequential actions behind human approval. This was the operating layer that later informed the Paperclip organization design.

  • specialized operating roles
  • cross-agent memory
  • human approval boundaries
  • private network access

Hermes · Honcho + pgvector · OpenWebUI · Telegram · Cloud SQL · Tailscale

chickenbuttjerky.com
BuiltEnterprise Mobility

Clippy Research Concierge

A guarded front door for intake, status, decisions, and deliverables

Built an internal OpenWebUI concierge that turns a plain-language request into governed research intake, reports status, surfaces pending decision cards, and retrieves deliverables. Its board tools are narrowly scoped: Clippy can relay Ben's explicit decision, but it cannot make that decision.

  • plain-language intake
  • board-scoped tools
  • decision cards
  • no delegated authority

OpenWebUI · Paperclip · MCP · identity-aware tools

IntegratedEnterprise Mobility

Figma Research Path

Board-scoped design analysis with truth-in-labeling

Integrated a Figma research capability that recovered exact text, spatial layout, and connector flow and wrote attributed comments during evaluation. The proven scope is reading, layout recovery, connectors, and comments; complete canvas generation and unrestricted write automation remain outside the claim.

  • exact-text recovery
  • spatial layout
  • connector flow
  • attributed comments

Figma API · Paperclip · MCP · board-scoped access

ConfiguredEnterprise Mobility

Professor CCC Research Organization

A four-role, citation-gated organization provisioned as code

Configured a research organization for Car Connectivity Consortium documentation with roles for research, specification mining, writing, and adversarial claim challenge. The organization is provisioned and review-gated; it is labeled Configured because a completed research dossier is not yet evidenced.

  • 4 configured roles
  • citation gate
  • adversarial challenge
  • no output claim

Paperclip · config as code · structured citations

BuiltChicken Butt Jerky LLC

Grant Sniper

Evidence-first grant discovery with a hard human submission boundary

Built SCOUT and WRITER workflows that discover opportunities, apply deterministic eligibility gates, deduplicate results, and assemble filing kits with source evidence. The system is intentionally draft-first: it prepares research and application material, but a person retains the final review and submission decision.

  • deterministic eligibility gates
  • deduplicated opportunity set
  • evidence-backed filing kits
  • human-only submission

Hermes · Python · public grant sources · structured evidence

BuiltPersonal R&D

Local AI Cost Tracker

A no-dependency view of multi-tool AI usage

Built a local dashboard that normalizes session, token, and estimated-cost records across Claude Code, pi, Codex, and OpenCode. It gives builders one privacy-preserving place to inspect usage patterns without sending the underlying development history to a hosted analytics service.

  • 4 supported tool families
  • local processing
  • zero runtime dependencies

TypeScript · local browser UI · CLI data adapters

BuiltEnterprise Mobility

CCC Intelligence Hub

Document intelligence that collapses weeks into hours

A 7-stage pipeline that scrapes a gated external portal, enriches every document with AI summaries and extracted action items, and serves it all as searchable knowledge, compressing weeks of manual review into a 2-hour auto-refresh cycle.

  • 254K+ chunks
  • 2,409 AI summaries
  • 4,763 action items
  • 2-hour refresh (from weeks)

Playwright · Python · Azure AI Search · PostgreSQL · Blob Storage

BuiltEnterprise Mobility

Call-Center Chatbot

Conversational AI with proven ROI

A pilot agent that lets support reps retrieve data and answer customer questions in natural language, surfacing answers in the conversation instead of hunting across systems. Direct proof of conversational AI tied to a measurable business outcome.

  • 40% shorter calls
  • +23% accuracy
  • +17% CSAT
  • measured pilot outcomes

RAG · LLM retrieval · Conversational UX

BuiltEnterprise Mobility

Cost-Cascade Model Router

Frontier capability at a fraction of the spend

A token-efficient router that opens every request on a cheap model and escalates to stronger, frontier-class models only when the task demands it, driving spend down without capping capability.

  • cheap-first escalation
  • up to frontier models
  • token-efficient by design

LiteLLM · multi-provider routing · Azure OpenAI + Foundry

BuiltEnterprise Mobility

Strategy Research-Agent Ecosystem

The six-specialist Hermes research fleet

Searches public sources and scrapes gated ones into a governed, trust-weighted RAG knowledge base with per-source allow/deny-listing, runs the analysis, and auto-generates decks, charts, podcasts, and short videos, with real-time voice plus STT/TTS. Delivered on OpenWebUI. As of 2026 this is the field-research layer under the automated research department.

  • 6 specialist Hermes agents
  • trust-weighted RAG
  • auto-generated media
  • real-time voice

OpenWebUI · Azure AI Search · STT/TTS · headless scraping

BuiltEnterprise Mobility

AI Coach Layer

Adoption engineered into the product

Three cooperating agents built on top of the research ecosystem: an avatar-and-voice onboarding trainer with a guided lesson library that spotlights the live UI, a passive on-demand coach, and a prompt coach that helps users improve requests to other agents.

  • 3 cooperating agents
  • guided lesson library
  • live prompt coaching

Azure VoiceLive · WebRTC · avatar + TTS

BuiltEnterprise Mobility

CarIntellect

Multi-model research with built-in conflict detection

A queue-driven pipeline that discovers and scrapes sources, then runs three independent AI analyses with explicit conflict detection, producing a citation-tracked, 14-section executive dossier with confidence scoring.

  • 3 AI analyses cross-checked
  • 14-section dossiers
  • citation-tracked
  • confidence-scored

Multi-model · Perplexity Sonar · PostgreSQL · PGMQ queue

BuiltEnterprise Mobility

Happy Robot

Autonomous outbound voice calls

Calls that bridge a realtime voice model to telephony, hold a natural 10-state conversation, and auto-chain a notification when a vehicle is ready, with custom audio transcoding between telephony and model formats.

  • 10-state conversation
  • realtime voice
  • telephony bridging

Azure OpenAI Realtime · Twilio Voice · FastAPI · WebSocket

BuiltEnterprise Mobility

CCC Voice Briefing Agent

A voice-first agent with a lip-synced avatar

Proactively monitors knowledge bases and briefs users out loud through a lip-synced avatar, with a fast-path design that answers in roughly two seconds and leader-elected monitoring across many indexes.

  • ~2s fast path
  • 8 knowledge bases
  • 12 search indexes
  • proactive monitoring

Azure VoiceLive · Foundry Agent · FastAPI · MCP · WebSocket

BuiltEnterprise Mobility

MCP Server Fleet & Factory

Standardized tooling for any agent

Containerized Model Context Protocol servers delivering ~45 tools (knowledge retrieval, voice, SQL analytics, live web scraping, presentation generation), consumable by any MCP-compatible agent, plus a standard factory lifecycle for building, deploying, and operating domain MCP servers with quality and telemetry gates.

  • ~45 tools
  • containerized servers
  • standard build/deploy lifecycle

FastMCP · Azure Container Apps · Azure AI Search

ArchitectedEnterprise Mobility

Enterprise AI Chat

One workspace for the whole org

An Entra-gated workspace giving teams frontier models, knowledge-base search, voice calling, slide generation, diagramming, and analytics through natural language, all in-tenant. The hub every other solution plugs into.

  • 5 frontier models
  • 6 knowledge bases
  • 35+ custom tools
  • 50+ neural voices

OpenWebUI · LiteLLM · Azure OpenAI + Foundry · FastMCP · PostgreSQL

ArchitectedEnterprise Mobility

Langflow Visual AI Workflows

Citizen-developer AI pipelines

A drag-and-drop builder where business users design multi-step AI pipelines and publish any flow as an MCP server, instantly turning a visual workflow into a callable tool.

  • 100+ components
  • 15+ model providers
  • flow-to-MCP publishing

Langflow · PostgreSQL · Redis · Celery · Bicep IaC

BuiltChicken Butt Jerky LLC

Clucky

A back office that runs on ~$180 a month

A first-generation four-model agent team supporting Chicken Butt Jerky's back office, ads, CRM, accounting, email, and content at a deliberately lean operating cost.

  • ~$180/mo
  • 7 MCP servers
  • 419 tools
  • first-generation stack

Multi-model agents · MCP · e-commerce + CRM + accounting integrations

BuiltChicken Butt Jerky LLC

Clucky Ears

An edge voice-and-vision kitchen terminal

A DFRobot ESP32-S3 AI camera turned into a hands-free voice-and-vision terminal: on-device audio capture with voice-activity detection streaming over a private Tailscale mesh to a Whisper-to-LLM-to-TTS bridge, with a camera path wired for a planned HACCP food-safety inspector.

  • on-device VAD
  • private Tailscale mesh
  • HACCP vision path

ESP32-S3 · C++/PlatformIO · Whisper · Node.js bridge

BuiltPersonal R&D

Ron Voice Clone

A fully local, on-device Hebrew voice clone

Hebrew text to Chatterbox TTS to RVC v2 conversion, running entirely on an M1 Mac. Rigorously evaluated: CER ~3% intelligibility and an ECAPA voice-match of 0.664, about the 48th percentile of the speaker's own real held-out recordings.

  • 100% on-device
  • CER ~3%
  • ECAPA match 0.664

Chatterbox TTS · RVC v2 · Apple Silicon

BuiltChicken Butt Jerky LLC

Breakout

AI-enriched GTM intelligence in one command

Turns raw distributor and field reports into a geographic, ICP-scored sales dashboard, enriching stores with trade-area and demographic data.

  • one-command pipeline
  • ICP scoring
  • geo + demographic enrichment

Python · SQLite · geocoding + enrichment APIs

BuiltEnterprise Mobility

Cloud Infrastructure Transformation

Static estate to demand-based, resilient infrastructure

Re-architected a static hub-and-spoke estate into a demand-based, auto-scaling model, retiring over-provisioned appliances for dynamic capacity and improving operational resilience.

  • 80% faster incident response
  • 45%+ lower opex
  • dynamic capacity

Terraform · Ansible · serverless functions

BuiltEnterprise Mobility

Reference Architectures & Guardrails

Governance by automation, not paperwork

The org's AI reference library: a zero-trust private-AI blueprint, an MCP factory pattern, an Agent Operations Center (LLMOps across runtime health, knowledge freshness, tool reliability, and change governance), and a platform-resilience blueprint. Guardrails that cannot regress: 19 hard invariants enforced by lint on every deploy, plus architecture decision records with revisit triggers.

  • 4 reusable blueprints
  • 19 lint-enforced invariants
  • targets 99.5% availability

LLMOps / AgentOps · IaC · policy-as-lint · ADRs

LedEnterprise Mobility

Innovation & Evaluation Framework

Turning ideas into funded, scored bets

A data-driven ideation engine that captures, scores, and prioritizes AI proposals on cost, KPIs, and business alignment, letting the department self-fund high-value bets. Paired with an enterprise API-security evaluation: a 140-requirement scoring matrix presented to senior leadership with a tested implementation plan.

  • 140-requirement scoring matrix
  • C-suite buy-in secured
  • self-funded portfolio

Portfolio scoring · vendor evaluation · executive advisory