How the work gets done.

Deep-dive project documentation — architecture decisions, what broke, how it was fixed, and what shipped. Each study covers the problem, the system design, and measurable outcomes.

Flagship · Live in Production

01 — EOR Compliance Checker (v5)

n8n · Azure OpenAI GPT-4o · HubSpot API · Custom JS rules engine
~242
deals auto-screened
103
mis-tagged deals corrected
46
nodes in workflow
238/238
clean writes before go-live

The Problem

Reps advanced deals to contracting and KYC without knowing if the role or industry was even eligible for EOR. Compliance only caught blockers at contract review — after the prospect was ready to sign. Lost deals, wasted KYC, eroded trust.

System Design

A 46-node n8n production workflow with three trigger paths: real-time webhook on deal creation/stage change, hourly safety-net sweep, and on-demand batch re-screen. The core architecture decision: the LLM only maps messy CRM text to one of 30 canonical sectors — a deterministic JavaScript matrix makes every tier decision. This means every verdict is repeatable, auditable, and cites its own reasoning.

Edge cases were reasoned about explicitly: pass-through detection (a staffing firm placing workers into oil & gas is screened as oil & gas, not staffing), semantic sector aliasing, and a "needs-title" hold instead of guessing on thin data.

"The core design decision — rules live in code, not in the model's head — is what makes it durable. The LLM does only what it's good at (reading unstructured text)."

What Broke & How It Was Fixed

Const-reassignment bug on first live test: caught immediately by the verification step; fixed and re-tested green. Prevention: the dry-run harness now gates every change.

Five analytics properties silently failed to write: didn't exist in the CRM. Folded content into the audit-note field instead. Prevention: validate property existence before writing.

Old prompt let thin data demote clean sectors: re-architected so confidence is a report field that never changes the tier.

📁 Project Documents

📄 Compliance Worksheet (30-sector, transcribed from official OHS Risk Guide)
📄 n8n Workflow JSON — 46-node production graph (NDA)
📄 Unit Test Suite — 16 tests covering sector mapping, tier assignment, edge cases
📄 Dry-Run Harness — report-only mode that tests rule changes against entire pipeline
📄 Go-Live Validation Report — 238/238 clean production writes
Live · Major Reliability Turnaround

02 — Post-Sales Call Automation

n8n cloud · Avoma webhook · Azure OpenAI · Azure Blob Storage (250+ country files)
21/25→0
failures eliminated
882→295KB
context bloat reduction
250+
country data files
4
AI agents in pipeline

The Problem

Reps spent time after every call on follow-up research and writing. The automation meant to handle it was failing 21 of 25 runs — a dead Azure AI Search index left four AI agents with no country context, producing generic filler instead of specific, grounded output.

System Design

Rather than recreate a fragile index (which needed permissions I didn't have), the whole failure point was designed out: extract countries mentioned in the call transcript → ISO-3 codes → fetch each file directly from Azure Blob Storage → inject into every agent. No fuzzy search, no credentials to rotate, graceful fallback so the email still sends if a file is missing.

"This is a textbook resilience rebuild. Replace the fragile service with a deterministic lookup — country code → exact file, no search index to die."

What Broke & How It Was Fixed

84% of failures: dead Azure AI Search index (DNS not-found): removed the dependency entirely; replaced with deterministic blob fetch.

Silent dead-ends on scheduled/future meetings: added meeting-completion guard + bounded retry budget (10→20 polls) + explicit exit nodes.

Broken paired-item lineage dropped transcripts downstream: carried the base object forward and converted 28 fragile item-references to a stable form.

📁 Project Documents

📄 Post-Sales Automation Close-Out Report (PDF)
📄 Country Context n8n Module — deterministic blob fetch architecture
📄 Failure Analysis CSV — 25-run diagnostic with root cause per failure
📄 Troubleshooting & Status Document
Shipped · Cost + Governance Win

03 — Clay Credit Governance & Cost Control

Clay API · Excel data modeling · Leadership one-sheeter
515K
credits traced to root cause
~50→~15
per-row credit reduction
3
build profiles shipped
7-sheet
forensic audit dataset

The Problem

Enrichment credit usage spiked from under 15K/month to 515K in a single month. 21% of spend ran through unattributed "Copy of" tables. Phone numbers were bought 2–3x per contact. A single day burned 201,924 credits from a phone-waterfall re-buying the same numbers.

System Design

Full-year forensic audit: a seven-sheet master dataset that re-attributed unnamed "Copy of" spend to real source tables. Then shipped a Standard Enrichment Table Template with three build profiles (Lean ~15 / Balanced ~21 / Full ~50 credits per row) and a credit-planning architecture: gate → cap → clone-don't-copy → dedupe → consolidate → weekly review.

📁 Project Documents

📊 Clay Credit Spend MASTER Data Package (Jun 2026) — 7-sheet Excel
📊 Clay Credit Spend Analysis (Jun 2026) — detailed Excel workbook
📄 Leadership One-Sheeter (PDF) — executive summary for VP Growth
📄 Detailed Report (Markdown) — 186-line narrative for Head of Revenue
📄 Standard Enrichment Table Template Design Doc
📊 Credit Usage by Table — June 2025 – June 2026 (CSV)
📄 Weekly Credit Review Governance Template
Civic Tech · Live Platform

04 — Your48 Civic Intelligence Platform

React · Vite · TypeScript · Mapbox GL · Chicago Data Portal API · Vercel
1,562
businesses catalogued
39
precincts mapped
$0/mo
infrastructure cost
4
data sources integrated

The Problem

Chicago's 48th Ward (Edgewater, Uptown, Andersonville) had no digital way for residents to find their precinct captain, look up local businesses, or submit service requests. The ward office managed 39 precinct contacts in Excel, 1,500+ business licenses as PDFs from the city, and service requests via paper forms. This doesn't scale.

System Design

Your48 V1.0 shipped as a static civic homepage with ward directory, city services, community resources, events calendar, and an AI chat widget. V2.0 adds precinct directory with interactive Mapbox map, searchable business license database from Chicago Data Portal, constituent intake forms with Airtable backend, and a public ward dashboard for transparency. The entire platform runs at $0/month on Vercel's free tier.

"Government technology shouldn't cost $50,000 in consulting fees. It should be fast, transparent, and nearly free to operate."

📁 Project Documents

📄 Project Brief — Your48 V2.0 executive summary & defense
📄 Architecture Diagram — SVG system design (5-layer diagram)
📄 PRD — detailed feature specs for precincts, businesses, intake, dashboard
📊 Seed Data — 1,500+ businesses from Chicago Data Portal API (JSON)
📊 Precinct CSV — 39 precincts with captain assignments
🗺️ Ward Boundary GeoJSON — precinct polygons for Mapbox overlay
📄 ROSTR Data Sync Pipeline — daily cron architecture
Published Research · Framework

05 — ROSTR Multi-Agent Framework

Published via Zenodo · DOI: 10.5281/zenodo.19550414
4
architectural components
5D
phase taxonomy
4D
priority scoring
3-tier
knowledge credibility

The Problem

Multi-agent AI systems suffer from context loss between sessions, naive task routing (keyword matching instead of phase-aware allocation), brittle retrieval (single-pass search without credibility control), and brittle prompting (raw prompts instead of compiled manifests). Production systems need architecture, not ad-hoc chains.

System Design

ROSTR formalizes four components: PAL (Prompt Abstraction Layer) compiles loose natural language into strict agent runtime manifests through a five-stage pipeline. RAG DAL (Dynamic Acquisition Layer) performs multi-pass retrieval across three credibility tiers (academic → editorial → community). NPAO classifies work into a 5D phase taxonomy (PreD → Design → Development → Deployment → Debugging) and scores priority on 4 dimensions. The Rostr Hub provides persistent state with four-level management (session → project → org → agent).

The framework is the backbone of the Gency AI agency — turning SOPs into intelligent agents — which cuts idea-to-scoped-project from 4–8 hours to under 30 minutes.

📁 Project Documents

📄 ROSTR Master Paper (PDF) — full academic publication
📄 CLAUDE.md — implementation guide with architecture, algorithms, templates
📄 PAL Compiler Specification — 5-stage intent compilation pipeline
📄 RAG DAL Algorithm — multi-pass retrieval with confidence scoring
📄 NPAO Taxonomy — 5D phase + 4D priority + agent allocation algorithm
🔗 Zenodo Record — zenodo.org/records/19550414
SDR Operations · Live Pipeline

06 — SDR-Agency Prospect Automation (AGS)

Clay · Amplemarket · ZoomInfo · HubSpot API · Custom JS Round-Robin Engine
5,398
contacts in master pipeline
188
verified HR contacts (competitive takeaway)
4
companies pulled per send (suppression)
100%
CRM-safe screening on every list

The Problem

A paid SDR agency needed a steady stream of clean, targeted, compliance-screened prospect lists every week — without contacting existing customers, open deals, closed-lost, or recently-touched contacts. Manual list building was slow, error-prone, and risked burning the agency's time (and the company's reputation) on dead-end or off-limits accounts.

System Design

Built a repeatable intake → normalize → dedupe → enrich → suppress → activate pipeline across Clay, Amplemarket, ZoomInfo, and HubSpot. Every company is screened live against the CRM before send: customer status, open deals, closed-lost, and active sequences are all checked. A custom JS round-robin engine assigns each prospect to the right sales pod based on territory (US West/Midwest/Northeast/South-Central + EMEA/Canada + Global).

The pipeline also includes an APAC filter that removes companies from Asia-Pacific markets using TLD detection (`.hk`, `.cn`, `.jp`, etc.), domain keyword matching, and company name analysis — catching APAC records even when HubSpot location fields are empty. Schools, governments, and competitors are filtered out via ICP screening.

"Treat prospecting as a governed data pipeline with source-of-truth boundaries — CRM = truth, enrichment = workspace, orchestration = glue. That's the difference between a list and a liability."

What Broke & How It Was Fixed

Parallel one-off workbooks caused drift: different team members built their own Clay tables. Consolidated into one master workbook with segment tabs as the operating source of truth.

APAC records slipping through with empty location fields: added domain TLD detection, domain keyword matching, and company name analysis as fallback detection layers — catching records that HubSpot location fields missed.

📁 Project Documents

📄 AGS_PROSPECT_AUTOMATION.json — 46-node n8n workflow with round-robin assignment engine
📄 Round-Robin Assignment Engine — territory-based pod routing (US West/Midwest/Northeast/South + EMEA + Global)
📄 APAC Filter v2 — multi-layer detection (TLD, domain keywords, company name, location fields)
📄 ICP Filter — school/government exclusion with NGO preservation logic
📊 5,398-Contact Master — segmented, tagged, CRM-safe
📄 Competitive Takeaway Play — 188 verified HR contacts at named competitor
Agency · Agent Development & Adoption

07 — Gency AI: SOPs to Intelligent Agents

ROSTR Framework · Claude Agent SDK · n8n · Make.com · Antigravity · Custom MCP
3-step
ROSTR-powered intake pipeline
4
core service verticals
24/7
agent operations
Live
accepting clients

The Problem

Organizations, SMBs, and consulting agencies have SOPs — documented processes that run their business — but turning those into autonomous AI agents requires deep expertise in agent frameworks, workflow orchestration, and platform configuration. Most teams don't have that in-house, and generic AI tools don't understand their specific business logic.

System Design

Gency AI uses a three-step ROSTR-powered process: JTBD Analysis maps the exact job the operator needs done, with deep intake to understand functional requirements. NLP Intent Extraction takes natural language descriptions and maps them to architecture. Then the ROSTR System delivers a full agent, workflow, or skill — production-grade automation the team can adopt instantly.

Service verticals span AI Agent Development (Claude Code, Codex, OpenClaw), Workflow Orchestration (n8n, Make.com), IDE & Platform Setup (Antigravity, Claude Code environments), and Web Apps & Custom MCPs that let agents interact with business logic. The interactive demo on the site lets prospects paste an SOP and watch the ROSTR engine generate architecture in real time.

"Turn any SOP into an intelligent agent. We empower operators to grow and scale their teams using customized workflows — from IDE setup to custom MCP integrations."

📁 Project Documents

🌐 Live Site — gencyai.vercel.app (5 pages: Home, Services, Process, Pricing, Work, About)
📄 Interactive SOP-to-Architecture Demo — ROSTR-powered wizard
📄 ROSTR Whitepaper — rostr-paper.vercel.app
📄 Calendly Booking — calendly.com/diamitani