Clay workflow

Linked tables for company qualification, clinician discovery, role validation, exact-domain matching, employment overlap, suppression, and campaign export.
I design the infrastructure behind modern outbound: ICP definition enrichment AI research signal-based targeting routing reporting
Built and operated systems generating 1,500+ ICP-qualified leads per quarter across data, automation, outreach, and deliverability.
A Clay-based GTM intelligence and qualification system that turned raw rehabilitation-company data into validated, suppression-checked, outreach-ready records.
Built a multi-stage Clay workflow for a referenceability-led outbound campaign targeting US outpatient physical, occupational, and speech-language therapy organizations.
The workflow separated company-level and person-level data, discovered current clinicians, validated roles and company associations using public evidence, matched prospects to known client contacts through exact company-domain logic, and calculated employment-period overlap before qualification.
It then applied suppression across customers, active pipeline, previous campaign batches, duplicates, invalid profiles, unverified roles, missing matches, and missing overlap before preparing personalized routing and campaign fields for Heddl.
Prevented clinic/company rows from being treated as individual people.
Returned structured person records with identity, role, company, evidence, and confidence.
Used candidate-domain = past-employer-domain instead of fuzzy company-name matching.
Required employment-period overlap before calling a prospect referenceable.
Kept raw-source tables separate from working, matching, and suppression tables.
Mapped routing fields separately from personalization fields for campaign execution.
A SuperAGI and n8n workflow designed to discover Product hiring signals, qualify hiring-manager context, and route reviewable records into the GTM review tracker without unsupervised outreach.
Designed a controlled hiring-signal campaign for approximately 164 target companies in the GTM review tracker. The workflow converted each company into a Product-hiring search query, processed companies in batches, discovered public job signals through SerpApi, normalized the response, removed duplicates, and validated Product relevance.
SuperAGI handled research, interpretation, and qualification. n8n handled deterministic operations including webhooks, batching, field normalization, duplicate prevention, payload mapping, and Google Sheets writes.
Qualified records were sent to a dedicated Qualified Hiring Managers tab for GTM review. Automatic outreach and BrandJet activation were intentionally disabled until job title, job URL, hiring-manager evidence, and contact quality could be validated.
Separated AI interpretation from deterministic workflow operations.
Used stable company, title, and job URL fields for deduplication.
Kept Target Companies and Qualified Hiring Managers as separate data layers.
Added GTM review as the approval boundary before any outreach activation.
Excluded BrandJet from the pilot to avoid acting on weak or false-positive signals.
Identified strict validation still needed for job title, URL, and job-linked evidence.
A closer look at the Clay research layer and the n8n orchestration layer behind the GTM systems I build. The motion is designed to make the workflow readable without turning the portfolio into a demo reel.

Linked tables for company qualification, clinician discovery, role validation, exact-domain matching, employment overlap, suppression, and campaign export.

Scheduled batch processing, job-signal discovery, normalization, deduplication, qualification, webhook handoff, and human review before outreach.
I can walk through the data model, qualification logic, automation design, and the decisions behind each system.
I design the layer between GTM strategy and execution: reliable data, automation, qualification, routing, outreach, and reporting.
Turn a business hypothesis into account lists, personas, segments, exclusions, qualification rules, and campaign-ready targeting logic.
Build structured company and person-level data using Clay, AI research, APIs, evidence URLs, validation, and deterministic matching.
Convert hiring, funding, technology, provider, and intent signals into prioritization, qualification, messaging, and timing decisions.
Connect enrichment, personalization, email, LinkedIn, WhatsApp, CRM, reply detection, and automatic stop conditions.
Protect campaigns with suppression, duplicate checks, role validation, domain authentication, bounce monitoring, and readiness gates.
Ship dashboards, SOPs, audit trails, Loom walkthroughs, and feedback loops so teams can operate and improve systems independently.
Two more PLF editions in build. Currently mapping 200+ companies across Mumbai and Chennai for expansion.
Six stages, automated end-to-end. You get a system you can run forever — not a campaign that dies when I leave.
No paid ads doing the heavy lifting. Just ICP-defined outreach across email, LinkedIn, and WhatsApp — for product leadership events attended by CXOs from Microsoft, Google, Salesforce, Dr. Reddy's, ADP, Lloyds.
Founders, attendees, and the people whose names land in inboxes I built.
Drop your name and company. See what an AI-personalized first-touch from one of my sequences looks like — generated live, in your browser.
// Real sequences inject 12+ data points (role, recent post, company news, mutual connections). This is the demo version.
Not "familiar with" — deployed in live campaigns generating real revenue.
// In your inbox within 24h. No spam. Just me.