Case Stories
Why the Claude gym hack matters for LatAm ops teams
Prompts are not firewalls. How a Bogotá clinic stopped their AI agent from hijacking patient triage rules by securing their Node.js/PostgreSQL API.
Kyto·Aug 13, 20260
The Journal
The Kyto Journal.
Latest
Zero AI hang-ups in one week: Inside the CDMX sales floor running GPT-5.6 Sol
How a 45-seat Mexico City outbound sales floor eliminated a 41% immediate hang-up rate by dropping legacy voice wrappers for raw API speed.
Read ArticleAll Articles
Case Stories
How we replaced an $8,000 clinic kiosk with OpenAI's $300 speaker
We swapped an $8,000 touchscreen for a $300 smart speaker running OpenAI's Realtime API. Patient check-in times dropped 50% and hardware costs fell 96%.
Kyto·Aug 9, 20260
Case Stories
How Circles boosted ARPU by 22% using a multi-agent OpenAI stack
Circles abandoned the single-prompt chatbot for a multi-agent architecture. The result? A 22% increase in average revenue per user and a 65% autonomous resolution rate.
Kyto·Aug 5, 20260
Case Stories
How a Medellín Operator Replaced Its WhatsApp Concierge With Alexa Plus
A 20-unit property group in El Poblado fired its digital front desk, cutting recurring support payroll to zero by integrating Alexa Plus directly into the rooms.
Kyto·Jul 24, 20260
Case Stories
How LatAm B2B sales teams use ChatGPT Work to automate meeting prep
A Bogotá SaaS firm replaced 15 hours of manual meeting prep per week with an autonomous AI agent. Here is how they doubled active selling time and added $1.2M in pipeline.
Kyto·Jul 16, 20260
Case Stories
How a 60-person Monterrey ISP automated tier-1 support using Deutsche Telekom's AI playbook
We dropped customer response times from 45 minutes to exactly 12 seconds for a regional ISP by wiring an LLM directly to their WhatsApp, FreeRADIUS, and WHMCS databases.
Kyto·Jul 12, 20260
Case Stories
How a Monterrey distributor copied HP's OpenAI architecture using n8n
How a 150-person B2B distributor in Mexico bypassed massive software costs to automate 4,000 weekly logistics updates, saving 62 hours a week.
Kyto·Jun 30, 20260
Case Stories
How to copy Omio's AI booking playbook on WhatsApp
Mateo’s operations team used to wake up to 45 unread WhatsApp messages. Today, they wake up to $3,200 in cleared reservations with zero human intervention. Here is the exact playbook.
Kyto·Jun 26, 20260
Case Stories
Why State Farm's AI commission cut matters for LatAm sales floors
A 215-person SaaS firm in Bogota reduced CAC by 28% by rewriting commissions for AI-assisted sales. Here is why the old commission playbook is officially a liability.
Kyto·Jun 22, 2026
Case Stories
How BBVA pushed ChatGPT to 120,000 employees (and what your operations team can copy)
BBVA didn't hit 80% daily active usage on ChatGPT by sending a PDF manual. They built a behavioral machine. Here is how they saved 3 hours per employee every week.
Kyto·Jun 18, 20260
Case Stories
The $20k Local Dev Shop Trap (And Why a Bogotá Ops Manager Bypassed It in 14 Hours)
Diego canceled a $20,000 IT contract on a Friday and built the software himself by Sunday using an AI agent. Here is how LatAm ops teams are bypassing the IT backlog.
Kyto·Jun 10, 2026
Case Stories
How a 50-person Bogotá brokerage copied Travelers' AI claims pipeline for pennies
Seguros Andinos replaced a 72-hour manual WhatsApp bottleneck with an automated n8n and GPT-4o pipeline, slashing claim processing times by 60%.
Kyto·Jun 6, 20260
Case Stories
What OpenAI's self-improving tax agent means for LatAm finance ops
In six weeks, an accounting back-office went from manually reviewing 7,000 tax returns to watching an autonomous agent write its own code patches. Here is the exact breakdown.
Kyto·May 29, 20260
Case Stories
How a LatAm Dental Network Copied AdventHealth's AI Pipeline Using n8n
A 20-location dental clinic in Mexico City cut patient chart reviews from 10 minutes to 2 minutes by replacing manual WhatsApp entry with a self-hosted n8n and OpenAI pipeline.
Kyto·May 25, 2026
Case Stories
How we built an AI invoice pipeline to eliminate manual SAP data entry
See how a 150-person distribution company in CDMX cut invoice processing from 7 minutes to 12 seconds using n8n and GPT-4o.
Kyto·May 17, 2026
Case Stories
How a $10M Bogota brand automated 90% of WhatsApp order tracking
A Bogota apparel brand was drowning in 1,800 monthly 'where is my order' tickets. By connecting Shopify directly to WhatsApp via API, they cut manual tracking tasks by 90% and reduced response times from 4 hours to 3 seconds.
Kyto·May 13, 20260
Case Stories
How a Cartagena Travel Agency Tripled Bookings with n8n and AI
CaribeNav dropped their WhatsApp response time from 3 hours to 12 seconds by automating custom quotes and availability checks for just $185 a month.
Kyto·May 9, 2026
Case Stories
How a Bogotá real estate agency cut WhatsApp response times to 2 minutes
By moving from manual phones to AI qualification, Inmowa cut lead response times from 4.2 hours to 1.8 minutes and increased property tours by 33%.
Kyto·May 5, 2026
Case Stories
How Choco automated 8.8M messy orders without changing client behavior
Choco stopped trying to force LatAm chefs to use B2B portals. By using OpenAI to ingest chaotic WhatsApp voice notes and blurry photos, they doubled sales productivity.
Kyto·May 1, 20260