Santiago Aday. Start a build →

GTM engineerRevenue systemsAI agents

I build the go-to-market machine.

Salesforce, the website, the ad platforms, and the AI agents that run between them. I turn sales, marketing and onboarding into one measurable system, then put AI to work inside it.

Taking one outside build at a time

SANTIAGO ADAY
GTM engineering · Miami, FL

  • AI messages sent45K+
  • CRM data points enriched270K+
  • Meetings scored by AI13,600+
  • Empty slots refilled~1,000
  • Lost accounts re-engaged6,100+
  • Accounts enriched nightly12K
  • Hours of manual work removed640+
  • AI agents live15

Software cut or avoided$130K+ / yr

Every number has a receipt
45K+
AI-written emails and texts sent
270K+
CRM data points enriched by AI
13,600+
sales meetings scored by AI
640+ hrs
of manual work taken off the team
$130K+
a year of software replaced or avoided
15
AI agents in production
01 / The machine

One system, from the first click to the renewal.

Most go-to-market stacks are six tools that don't talk to each other. I build the connective tissue, in the CRM, on the website and inside the ad platforms, and I put agents on the work nobody has time for.

01

Traffic

Ads and search

  • Ad platforms fed with real revenue
  • Ad-spend audits and monitors
  • Attribution you can reconcile
02

Capture

Site, pages and forms

  • Faster landing pages
  • A shorter lead form
  • Tests on every release
03

Meeting

Booked to held

  • AI reminders and texts
  • Risk scoring on every booking
  • Empty slots refilled
04

Deal

Pipeline and follow-up

  • One action list per rep
  • CRM filled in from calls
  • Same-day follow-up drafts
05

Onboard

Kickoff to live

  • Customer calls analyzed
  • Recaps sent automatically
  • Prep done before the first call
06

Retain

Risk, renewal, win-back

  • Risk and churn signals
  • Usage on every account
  • AI win-back outbound
And the loop closes: meetings held and deals won flow back to the ad platforms, so bidding learns from revenue.
02 / Selected work

Eleven systems in production.

Each one has the problem, what I built, the result, and the method behind the number. All of it shipped in-house at a venture-backed B2B SaaS company.

01 AI outboundAI-written outbound that reopens dead pipelineSalesAI
7,600+lost and churned contacts re-engaged
+73%reply rate vs the old generic sequence
44dead deals reopened

Personal first-touch emails to every new lead, and win-back sequences to lost deals and churned accounts.

The problem

New leads got a generic sequence, and lost deals and churned accounts got nothing at all.

What I built

AI email lanes that sit on top of the sales engagement tool. One writes the first personal email to every new lead who hasn't booked. Others write win-back outreach from the account's own history: why they left, what they used, what has shipped since. Every send is logged with replies and outcomes, and variants are assigned at random so results can be trusted.

Results

  • 7,600+ lost and churned contacts across 6,100+ accounts re-engaged
  • Reply rate up 73% against the old generic sequence
  • 44 deals reopened and 16 won back
  • 2,200+ AI first-touch emails to new leads, 90 of whom booked a meeting

Stack

n8nSalesforceOutreachClaudeGPT

ReceiptReply rate is an early read against the previous sequence. Deals counted within 75 days of enrollment.

02 Meeting remindersAI that gets booked meetings to actually happenSalesAI
35K+AI-personalized reminders sent
+3 ptsshow rate vs generic copy
~1,000empty meeting slots refilled

AI reminders, texts that read replies, and automatic slot refills on every booked sales meeting.

The problem

Booked meetings were no-showing or cancelling at the last minute, and the empty slots were wasted rep time.

What I built

An engine that writes a personalized reminder for each prospect, sends morning-of emails and texts, reads the replies, and puts freed slots back on the calendar. The AI copy was tested against a generic version on a random split.

Results

  • 35,000+ personalized reminders across nearly 16,000 meetings
  • AI-written reminders beat generic ones by about 3 points of show rate
  • Nearly half of texted prospects reply, and those who can't make it say so hours ahead
  • About 1,000 freed slots refilled with new bookings

Stack

n8nSalesforceSMSClaudeGPT

ReceiptMeasured on a randomized split of 2,266 meetings.

03 CRM autofillAI that fills in the CRM so reps don't have toSalesRevOpsAI
270K+CRM data points enriched by AI
117fields kept current automatically
330+ hrsof manual data entry removed

Reads sales calls and writes the answers into the CRM.

The problem

Reps left CRM fields blank, so forecasting, routing and reporting ran on missing data.

What I built

An agent that reads every sales call and writes what it learned into the CRM, with guardrails so it can only touch fields that have been approved.

Results

  • 270K+ data points written or updated in under five months
  • 117 fields kept current with no rep effort
  • 330+ hours of manual data entry removed

Stack

GongSalesforceClauden8n

ReceiptHours are counted as if done by hand, at five seconds a field.

04 Meeting scoringAI that tells reps which meetings won't happenSalesAI
13,600+meetings scored for risk
5xearlier notice on bad meetings
128 hrsof rep calendar time handed back

Scores every booking for risk and frees rep calendars hours earlier.

The problem

Reps were spending prime calendar time on meetings that were never going to happen: bad-fit leads, duplicates and no-shows.

What I built

A scoring engine with a rule layer plus an AI auditor. It scores each booking, explains its reasoning, and recommends which meetings to cancel so the slot can be reused.

Results

  • 13,600+ meetings scored
  • Flagged meetings were less than half as likely to happen
  • Reps get 16+ hours of notice on a bad meeting instead of about 3
  • 128 hours of rep calendar time handed back, most of it refilled

Stack

n8nSalesforceSlackClaude

ReceiptOutcomes counted only for meetings left on the calendar.

05 Rep operating systemOne prioritized action list for every repSales
225K+calls, texts, emails and meetings in one ledger
100%of open leads visible, up from 11%
2 weeksfrom first release to reps using it

A CRM home page that tells each rep who to call next, backed by one ledger of every customer touch.

The problem

Speed-to-lead and follow-up weren't measured anywhere. Activity lived in separate tools, and reps had no single list of what to do next.

What I built

An operating layer for sales inside the CRM: one ledger fed by every dialer, sequencer and text, speed-to-lead clocks, a recovery path for no-shows, and a home page with one prioritized list and one-click outcomes.

Results

  • 225,000+ calls, texts, emails and meetings unified into one ledger
  • Reps can now see 100% of their open leads, up from 11%
  • In daily use by the first sales team two weeks after the first release
  • Every dashboard number reconciled to a raw recount

Stack

SalesforceApexLWCZoom PhoneOrum

ReceiptLedger reconciled record by record against the source systems.

06 Paid mediaAd spend that optimizes for revenue, not form fillsMarketingRevOps
$3Mof ad spend audited
5ad platforms fed with real revenue
100%of leads the ad platform can match, up from 6%

Closed revenue flows back to the ad platforms, and a monitor catches spend chasing the wrong signal.

The problem

The ad platforms were bidding on page views and form fills, not on meetings held or deals won, and cost per lead had doubled without anyone knowing why.

What I built

An engine in the CRM that sends meetings held and deals won back to every ad platform, with delivery checks so nothing is silently dropped. Alongside it, a full audit of the search account that traced the cost increase to its cause, and a monitor that flags when bidding starts chasing a low-intent signal.

Results

  • Found $3M of spend optimizing toward a signal that wasn't revenue
  • Revenue feedback built for five ad platforms
  • 100% of leads now carry the data an ad platform needs to match them, up from 6%
  • 23 brittle automations replaced by one engine

Stack

SalesforceGoogle AdsMetaMicrosoft AdsPostHog

ReceiptA finding and a fix, not recovered dollars. No bidding lift is claimed yet.

07 Lead funnelA faster lead funnel that can't ship brokenMarketingPlatform
3 to 1steps to book a meeting
250 / moleads that regained their source
1,000+automated tests on every release

A shorter form, faster landing pages, and automated tests on every release.

The problem

The lead form was slow and three steps long, landing pages were heavy, and nothing checked the funnel before a change went live.

What I built

A one-step form straight into the scheduler, landing pages rebuilt for speed, attribution that survives cookie blockers, and a browser test suite that blocks any release that would break the path from ad click to booked meeting.

Results

  • Lead form cut from three steps to one
  • 150+ landing pages rebuilt for speed
  • About 1,000 automated tests on every release
  • Attribution restored on about 250 leads a month

Stack

Next.jsWebflowPostHogHubSpotPlaywright

ReceiptNo conversion lift is claimed. None has been measured yet.

08 AI teammateAn AI teammate the whole revenue team can askRevOpsPlatformAI
70tickets shipped end to end in 4 weeks
16systems it can read and explain
10,000+fields, flows and workflows catalogued

One AI-readable map of every GTM system, with an agent on top that answers questions and ships changes safely.

The problem

Every data question and every small change needed an engineer, and nobody had one view of how the stack fits together.

What I built

A living catalog of every field, flow and workflow across the stack, a glossary that maps business terms to exact fields, and an AI teammate on top. Anyone can ask it a question. Builders can have it ship changes, with approvals, backups and limits on what it may touch.

Results

  • 70 tickets delivered end to end by the agent in its first four weeks
  • 16 GTM systems in one AI-readable map
  • 10,000+ fields, flows and workflows catalogued and kept in sync
  • Permission tiers, so AI can only do what each person's role allows

Stack

Claude CodeMCPTypeScriptSalesforceHubSpotn8n

ReceiptCounts come from the synced catalog, not estimates.

09 Tool consolidationSoftware you stop paying for, rebuilt in the systems you ownCX and onboardingRevOps
$130K+a year of software replaced or avoided
22,000+records migrated with zero data loss
12Kaccounts enriched nightly

When a tool's job can be done better in the CRM or by an agent, I build it and migrate cleanly.

The problem

The stack had grown one tool at a time. Data was split across systems, renewals kept arriving, and several tools did jobs the CRM or an agent could do better.

What I built

A customer success workspace inside the CRM, sales sequences with AI steps, and an agent that took over list imports, dedupe and enrichment. Each one started with an audit of what the old tool really did, so only what mattered was rebuilt.

Results

  • $130K+ a year of software replaced or avoided
  • 22,000+ records migrated with zero data loss, verified value by value
  • 12,000 accounts enriched every night with product-usage signals
  • Tools evaluated and turned down when the evidence didn't support them

Stack

SalesforceApexSnowflakeIntercomClaude

ReceiptReplaced or avoided: some tools are already off, some retire at renewal, some were never bought.

10 Customer-side agentsAI that logs, recaps and prepares the customer workCX and onboardingAI
2,600+customer calls analyzed
2xmore customer sessions logged
~180 hrsof manual updates removed

Agents that analyze customer calls, update the CRM, send recaps and capture churn signals.

The problem

Customer sessions weren't being logged, recaps were written by hand, and churn signals sat in chat threads.

What I built

A set of agents for the onboarding and support teams, built with a teammate: one reads each customer call and updates the CRM, one sends the recap, one captures cancellation requests from chat, and one prepares the first call.

Results

  • 2,600+ customer calls analyzed
  • Session logging doubled
  • 700+ customer recaps sent automatically
  • 270+ cancellation requests captured without manual entry

Stack

ClaudeGPTGongSalesforceIntercom

ReceiptBuilt with a teammate on a shared agent platform.

11 Engineering disciplineGTM systems that ship like softwarePlatformAI
350+changes shipped in four months
4 to 1test runs to reach production
96%less alert noise

Version control, tests, reviews and monitoring for the revenue stack, so changes are fast and nothing breaks quietly.

The problem

Changes to the CRM and the marketing stack went out by hand, with no review, no tests and no record of what changed when a metric moved.

What I built

A release pipeline for the CRM with test gates, an AI code reviewer and human approval for production. A monitoring service that watches the website, ads and booking flow for changes. And production errors that turn into tickets on their own.

Results

  • 350+ changes shipped through the pipeline in its first four months
  • Production validation cut from four test runs to one
  • Alert noise cut 96%, with every real alert kept
  • Production errors ticketed and fixed automatically

Stack

GitHub ActionsSalesforce DXAWSTypeScriptClaude

ReceiptAI writes and validates in a sandbox. A human approves every production change.

03 / Agents

Fifteen AI agents doing real work, every day.

They read calls, write CRM data, personalize every reminder, score risk and draft follow-ups. Each one is registered, costed per run, and measured against a control wherever a control is possible.

01

Meeting reminders

Writes a personalized reminder email and text for every booked meeting.

+3 pts show rate
02

Reply reader

Texts prospects before meetings and reads the replies, so reps know who can't make it.

Nearly half reply
03

Meeting scoring

Scores each booking for risk and recommends which to cancel.

5x more notice
04

CRM autofill

Reads sales calls and fills approved CRM fields.

270K+ data points
05

First-touch email

Writes the first personal email to every new lead who hasn't booked.

2,200+ emails sent
06

Win-back outbound

AI-written outreach to lost deals and churned accounts.

+73% reply rate
07

Follow-up drafts

Drafts the follow-up from the call transcript and CRM data.

90% land same day
08

Call analyzer

Reads each customer call and updates the CRM.

2,600+ calls analyzed
09

Session recap

Sends the customer a recap after each session.

700+ recaps sent
10

Churn intake

Reads cancellation requests and logs the reason and outcome.

270+ captured
11

First-call prep

Builds the deck before a customer's first call.

Ready in under 3 hours
12

Lifecycle emails

Account-aware customer emails written by AI, tested in shadow first.

12 AI-written steps
13

Inbound booking

Reads appointment emails and books the meeting automatically.

No manual entry
14

Inbox responder

Drafts replies in the sender's own voice, behind an approval step.

Human-approved
15

Change summaries

Explains every system release in plain English.

Every release
04 / Proof

Measuring AI honestly.

Every agent number I publish goes through the same four tests.

  1. 1The result has to come after the AI acted.
  2. 2The comparison has to be fair: a real before, or a control group.
  3. 3If the AI chose who got treated, it doesn't count as proof.
  4. 4Time saved is labelled as an estimate, never booked as revenue.

Where I can, I run randomized splits against a control group. Numbers that don't survive review get withdrawn, and five were. That discipline is why the ones on this page hold up in front of a skeptical CFO.

05 / Revenue leaks

Finding where revenue leaks.

Most go-to-market stacks lose money quietly: ad spend that optimizes for the wrong thing, data that stops syncing, leads nobody sees. I find the leak, prove it with data, and fix it.

The symptomWhat it wasOutcome
The symptomAd spend chasing the wrong signal
What it wasA page view was counted as a conversion, and automated bidding optimized toward it for months.
$3M audited
The symptomRep activity missing from the CRM
What it wasCalls had silently stopped syncing for two dozen reps, with no error anywhere.
24 reps restored
The symptomLeads that reps never saw
What it wasThe work queue hid open leads with missing dates or stale timers.
11% visible, now 100%
The symptomFake signups in the pipeline
What it wasFingerprinted the bot and found the one control that stopped it without blocking real buyers.
1,800+ fake leads removed
The symptomConversions the ad platform never counted
What it wasThe API said OK on receipt and rejected the data later.
Caught and fixed
The symptomA performance gap between reps
What it wasIt was routing, not performance. Once deal size was controlled, the gap disappeared.
Settled with statistics
The symptomTwo dashboards, two answers
What it wasBoth were right. They counted cancellations differently, so I reconciled them to one rule.
One number
06 / Software retired

Buying less software.

Most stacks pay for tools whose job the CRM or an AI agent can now do better. I rebuild that job inside the systems you already own, migrate the data cleanly, and you stop paying the renewal.

  • Customer success platformRebuilt inside the CRM
    $60Ka year
  • Sales engagement platformCRM-native sequences with AI steps
    $45Ka year
  • Data tools, add-ons and automationsHandled by AI agents and in-house apps
    $25K+a year
A year, replaced or avoided$130K+
07 / Stack

Deep in the tools you already run.

I work in the tools themselves and in the code around them: Apex and Lightning, TypeScript and Python, CI pipelines and agent tooling.

Salesforce

1,100+ custom fields, 250+ Apex classes, 100+ flows and 39 Lightning components shipped.

HubSpot

Marketing automation, nurture and segment workflows, and a full property catalog for AI.

AI agents

Claude and GPT in production, MCP servers, tracing, and cost logged per agent run.

n8n and automation

The engines behind the AI programs. Production errors become tickets automatically.

Outbound and telephony

Sequencers, dialers and SMS, deduplicated into one ledger of every touch.

Paid media

Google, Meta, Microsoft, X and ChatGPT Ads, with attribution reconciled to the leadership dashboard.

Website and forms

Webflow, Next.js and Calendly: landing pages, lead forms, routing and scheduling.

Analytics and experiments

PostHog with identity that survives cookie blockers, A/B tests and funnel dashboards.

Support and success

Intercom, NPS routing, risk scoring, and a two-way bridge into the CRM.

Calls and data

Call transcripts turned into CRM data, and a nightly warehouse sync onto every account.

Enrichment and lists

Apollo and Clay for enrichment, with imports, dedupe and tagging handled by an agent.

CI and cloud

GitHub Actions, AWS and Playwright: release gates, monitoring and automated browser tests.

08 / Kind words

From the people who use it.

“The texts that are going out on our behalf have been super clutch and have saved me plenty of time, effort, and peace of mind.”
Account Executive on AI meeting texts
“Wow dude this is INSANE! Really great job, everyone is talking about it now.”
Marketing leader on the attribution report
“That's absolutely amazing.”
GTM operations executive on an intake automation
“Huge lift!”
Sales leader on the AI win-back sequence
“Love that you asked for explicit approval. This is approved.”
Marketing lead on shipping with guardrails

09 / Work with me

Have a system that's leaking?

I run GTM engineering in-house and take one outside build at a time. The first build ships in about three weeks at a fixed price, and your team owns it afterwards.

Email me →santi [at] santiagoaday.com
  1. 01A 30-minute call

    You show me the system that's leaking.

  2. 02A fixed scope and price

    In writing, before anything starts.

  3. 03About three weeks

    Built, tested and in production.

  4. 04A documented handoff

    Your team owns it from day one.

Builds I take on

AI outbound that sounds like you

Win-back, first-touch and follow-up emails written from each account's own history, logged and measured.

An AI agent inside your CRM

Reads calls, fills fields, enriches records or scores risk, with guardrails and a measured result.

Meetings that actually happen

Reminders, texts and risk scoring on every booking, with empty slots refilled automatically.

Ad platforms that learn from revenue

Won deals sent back to the ad platforms, so bidding stops chasing form fills.

One action list per rep

Speed-to-lead clocks, a single ledger of every touch, and a home page that says who to call next.

A tool you no longer need

The job a vendor does, rebuilt inside the systems you already own, with a clean migration.