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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.