Headcount Zero

Jordan
  • 0,0 (0)
  • Empreendedorismo
  • Semanalmente

Headcount Zero is about scaling a company with AI and no employees. Each week, Jordan breaks down one practical system to win customers, deliver better work, and protect margin.

  1. -2 h

    Build a Trust Center That Turns Security Questionnaires Into 40-Minute Tasks

    After losing a $14K/month deal because he couldn't answer a 48-page security questionnaire, Jordan built a system that turns vendor security reviews from deal-killers into 40-minute tasks. In this episode, he walks through building a public Trust Center in Notion, creating a reusable answer library mapped to CAIQ-Lite and SIG Lite standards, and automating questionnaire responses with a Make flow that handles 70% of questions automatically while routing the rest for human approval. Jordan covers the two industry-standard "lite" frameworks most mid-market buyers accept, shows exactly which artifacts to publish (SOC 2 links, DPA templates, subprocessor lists), and explains how to triage questionnaires into three risk tiers so you only escalate when truly necessary. He also addresses when this approach breaks down—regulated data scenarios that require full SIG Core responses—and how to charge for deeper reviews as a line item rather than free pre-sales work.

    16 min

  2. -2 dias

    Production RAG That Actually Works: Supabase pgvector with Bounded Corpus and Auto-Refresh

    Most private RAG deployments fail not because of the model, but because of three fixable infrastructure problems: stale embeddings from unchanged source documents, unbounded corpora that contaminate results with irrelevant content, and no retrieval guardrails to catch bad matches before they reach the LLM. This episode walks through building a production Supabase pgvector RAG system that solves all three: automatic re-embedding when source documents change, bounded corpus design, and SQL-level retrieval gates that block the LLM call when similarity scores are too low or keyword overlap is insufficient. You'll get the complete schema, indexing strategy, refresh pipeline using Make.com or n8n, and the exact cost breakdown for running this on Supabase Pro.

    19 min

  3. -3 dias

    Build a Regression Gate: Test Sets and Pass-Fail Thresholds for LLM Changes

    When a two-word prompt change can approve fourteen spam submissions in one weekend, you need more than eyeballing outputs. This episode builds a regression gate—a frozen test set, three types of automated checkers, and a pass-fail threshold that blocks deploys when something breaks. The whole system runs inside Promptfoo, triggered from Make or n8n, and takes about forty-five minutes to set up from scratch. You'll get the exact CLI commands, environment variables, and exit code logic to catch prompt regressions before your clients do.

    16 min

  4. -3 dias

    Why You Don't Need to Self-Host LLMs for Enterprise Clients

    Most solo operators assume they need to self-host LLMs when enterprise prospects mention "on-prem" requirements. But you can satisfy 90% of enterprise security questionnaires without touching a single GPU driver. Jordan walks through the four controls that replace self-hosting: BYO API keys per tenant, region-pinned endpoints, structured logging with correlation IDs, and a visible kill switch in the client portal. You'll get the exact provider settings for Azure, OpenAI, Vertex AI, and Anthropic, plus a one-page checklist you can send to procurement the same day you get the ask.

    16 min

  5. -3 dias

    Build a Client Control Surface with Airtable + Softr

    When clients text you at midnight asking to turn off automations, you've hit the client proof problem. This episode walks through building a minimal control surface—not a full portal—that gives clients on/off toggles, SLA status chips, and a panic switch using Airtable + Softr + Make. Total build time under a day, total cost under $140/month. You'll get the exact schema, webhook wiring, and role-based visibility setup, plus a Notion + Super fallback for read-only clients. No custom code, no complex auth—just proof and control for the two things clients lose sleep over: is it working, and can I stop it if it's not.

    15 min

  6. 18/05

    Your 60-Second Rollback Plan for No-Code Workflows

    Solo automation consultants running Make, Zapier, or n8n in production face a hidden risk: every platform has version history, but each has gaps that will cost you time, money, or client relationships during an actual incident. This episode delivers semantic versioning for no-code workflows, Git-backed exports that work on free tiers, and practiced 60-second rollback drills for each platform. Jordan walks through the exact system that saved him from a costly HubSpot sync failure and shows you how to build rollback-on-demand into your client automation practice.

    17 min

  7. 15/05

    Build Tamper-Evident Run Logs: One ID, Full Audit Trail

    Learn to build a complete data lineage automation system using correlation IDs, append-only logs, and exportable evidence. Jordan covers three implementation stacks - Airtable with permissions and interfaces, Notion with integration-only writes, and Postgres/Supabase with hash-chained tamper evidence. Plus a Slack slash command that answers "what happened on this run" with timestamps and status codes in under three seconds. Turn invisible automation work into defensible, auditor-ready evidence that shifts client conversations from "trust me" to "here's the proof."

    14 min

  8. 13/05

    Build a PII Redaction Subflow That Logs Every Detection

    Solo operators running Make or n8n workflows that touch client data face a hidden risk: PII and secrets leaking through LLM calls, API requests, and workflow logs. While vendors offer inconsistent protection—Vertex AI has non-configurable filters, Bedrock Guardrails cost per request, and Anthropic relies on prompt patterns—you need a solution that works everywhere and proves it's working. This episode walks through building a detect-mask-log subflow that drops in before any external call. Using Microsoft Presidio or OpenAI's new Privacy Filter, you'll scan for emails, phone numbers, API keys, and other sensitive data, then mask or pseudonymize it based on rules you control. The real game-changer is the annotated log—a JSON object that records what was found, what was masked, and which detector version ran. Jordan covers the cost math between self-hosted detection ($5-15/month in compute) versus managed services (AWS Comprehend's per-character billing, Google DLP's per-GB pricing), addresses the quality tradeoffs of masking versus pseudonymization, and shows how to fail closed when detection errors occur. You'll get the exact subflow templates for Make and n8n, plus the log schema that turns invisible security work into client-visible protection.

    16 min
Ver tudo (27)

Sobre

Headcount Zero is about scaling a company with AI and no employees. Each week, Jordan breaks down one practical system to win customers, deliver better work, and protect margin.

Informação

  • Autoria
    Jordan
  • Anos em atividade
    2 mil
  • Episódios
    27
  • Classificação
    Para todos
  • Copyright
    © 2026 Headcount Zero
  • Site do programa
    Headcount Zero