Erwin Mongui

Engineering leader. Eighteen years of building the platforms behind customer-facing products, and the teams that run them.

Director of Technology at Cedar Planters. Certified AI Engineer. Based in Toronto.

See the work

I have spent 18 years turning ambitious roadmaps into systems that hold up in production: owning architecture end to end, from cloud migration and CI/CD to microservices and data models, and growing engineers through mentorship and structured career paths. I care about deployment velocity, technical debt, and whether the team is getting better every quarter. Since 2024, as a Certified AI Engineer, I have been bringing machine learning and AI-assisted development into that work.

Experience.

Five roles across four companies, from junior engineer in Bogotá to Director of Technology in the Greater Toronto Area.

January 2024 to present

Cedar Planters

Software Engineering Manager and Director of Technology

Oakville, Ontario

20% faster data retrieval and analysis after redesigning the data models
  • Lead cross-functional teams building scalable applications and own the full lifecycle, from requirements through deployment.
  • Redesigned the data models behind analytical and operational workloads, making retrieval and analysis 20% faster.
  • Run continuous improvements to the website, using A/B testing and third-party API integrations to lift performance and functionality.
  • Keep tracking pixels and scripts accurate so marketing works from complete data, and keep the site and applications secure and up to date.

May 2019 to November 2022

Fincaraiz.com.co

CTO and Software Engineering Team Lead

Bogotá, Colombia

50% lower infrastructure and maintenance costs after leading the cloud migration
  • Defined the technical strategy and owned architecture across several high-traffic, customer-facing products.
  • Led the migration from legacy infrastructure to the cloud, cutting infrastructure and maintenance costs in half.
  • Built continuous delivery pipelines with Jenkins and automated testing, so the team could deploy often while production got more stable.
  • Evolved the Python and Django microservice backend and launched 15+ REST APIs, reducing latency by 25%.
  • Hired, onboarded and grew engineers across backend and platform teams; metrics-driven prioritization raised delivery efficiency by 20%.

July 2015 to July 2018

Fincaraiz.com.co

Lead Software Engineer

Bogotá, Colombia

70% less manual operational work through the automation initiatives I delivered
  • Led the team that rebuilt the company's core customer-facing website, increasing user engagement by 25%.
  • Delivered automation that removed 70% of manual operational work.
  • Introduced Docker and Kubernetes, making deployments consistent and the platform more reliable.
  • Brought in Agile practices and coached junior engineers, improving delivery predictability and cutting operational costs by 25%.

May 2013 to June 2015

Periferia IT

Senior Software Engineer, consulting for Old Mutual and Editorial Legis

Bogotá, Colombia

50% sales growth in the first year of the payment service I built for Old Mutual
  • Built Old Mutual's payment microservice on MVC .NET, which drove a 50% increase in sales in its first year.
  • Delivered REST APIs on ASP.NET for Editorial Legis, improving content management efficiency by 40%.
  • Introduced automated testing that kept existing applications at 98% stability, and optimized SQL for a 30% performance gain on large data volumes.
  • Designed normalized relational schemas and wrote the complex SQL that turned large datasets into decisions.

February 2010 to June 2012

Americas BPS, for Hewlett Packard Enterprise

Junior Software Engineer

Bogotá, Colombia

15% productivity gain from the applications I designed and built
  • Designed and built C# and .NET applications that expanded system functionality and performance.
  • Shipped multiple projects on time, with the attention to detail and organization that kept production accurate.
  • Where the habit of writing well-documented, dependable code started.
Adevinta Culture Awards 2019 finalist digital diploma awarded to Erwin Mongui

2019

Adevinta Culture Awards finalist.

Recognized while I was CTO at Fincaraiz.com.co. The Culture Awards go to the people who build trust within their teams by living Adevinta's Key Behaviours:

  • Say it like it is, and treat feedback as a gift.
  • Win and lose together, and don't be a jerk.
  • Get stuff done and keep your promises.
  • Always be learning, fast. Try stuff, learn from the data, and try again.

How the numbers happened.

Three projects written up in full: the plan, the architecture, what went wrong and how the team got through it.

50%lower hosting cost, zero downtime

Monolith to microservices at Fincaraiz

Moving a real estate portal with 5M+ monthly visits from ASP.NET to Django, React and Kubernetes, and rebuilding search on the way.

Read the case study → 100%of leads delivered and reconciled

A lead distribution engine for any CRM

How a failed integration model became a Connector service that onboards partners in minutes, across six CRM platforms.

Read the case study → Months → hoursto produce a new report

A data platform from zero at Cedar Planters

As the first technical hire, replacing third-party reporting tools and spreadsheets with one platform the company owns.

Read the case study →
All case studies →

Measured in production.

A few of the numbers behind the roles.

18+

years building software and leading the teams that build it

2010 to today

50%

lower infrastructure and maintenance costs after a cloud migration

Fincaraiz.com.co

70%

of manual operational work automated away

Fincaraiz.com.co

25%

lower latency across 15+ REST APIs on a Python microservice backend

Fincaraiz.com.co

25%

more user engagement after rebuilding the core customer website

Fincaraiz.com.co

50%

sales growth in the first year of a new payment service

Old Mutual

AI, in practice.

Machine learning built on eighteen years of data-heavy platform work, and AI tools built into how the work gets done.

Certified Artificial Intelligence Engineer, Lambton College, 2024

Model building and training

TensorFlow and PyTorch for building and training models, with Pandas, NumPy and Matplotlib doing the data preparation and analysis that every model depends on.

TensorFlowPyTorchPandasNumPyMatplotlibPython

Data foundations for AI

Models are only as good as the data underneath them. At Cedar Planters I built the data platform and redesigned the data models serving analytical and operational workloads, making retrieval 20% faster. At Fincaraiz I led natural language, map-aware property search years before the generative AI wave.

Data architectureApache NiFiElasticSearchOpenSearchApache SolrPostgresMongoDB

AI-assisted engineering

Cursor and Windsurf are part of the daily workflow: scaffolding, tests, refactors and exploration, held to the same review and testing bar as any other code that ships.

CursorWindsurfVibe codingAI pair programming

Leading AI adoption

An engineering manager's view of AI: where it removes toil, where it needs guardrails, and how to know it is working. Metrics-driven prioritization and A/B testing decide what stays.

Metrics-driven prioritizationA/B testingDeveloper experienceEngineering guardrails

Notes from the job.

What the projects taught me about architecture, tools, AI and leading engineers.

Performance · October 8, 2026

Check before you believe

A headless demo against two real stores, 24 identical runs: why demos are fast and why one test is not a measurement.

 SEO · October 8, 2026

Best practices expire

Keeping SEO, GEO and AEO current with a dated log, a monthly AI review and a test before every change.

 Machine learning · October 4, 2026

A Shazam for movies, revisited

Our 2023 bot that named movies from a photo or a plot: three models, two vector indexes and no test set.

 Data science · October 4, 2026

51,000 ads, 190 rows

Revisiting our 2023 Kijiji visit model: a prediction that peeked at the answer, a rebuild, and what really drives visits.
All writing →

The stack.

Grouped by where it shows up in the work. Scroll sideways to browse.

Machine learning and AI

Certified AI Engineer, Lambton College, 2024. Model building in TensorFlow and PyTorch, data work in the Python stack.

TensorFlowPyTorchPandasNumPyMatplotlib

AI-assisted development

AI coding assistants as part of the everyday workflow, with the same review bar as any other code.

CursorWindsurfVibe coding

Languages

The everyday tools, with Python and C# carrying most of the backend history.

PythonC#JavaScriptTypeScriptSQL

Backend and architecture

Microservice platforms, REST APIs and the queues that keep them honest.

DjangoDjango-CeleryNode.js.NETREST APIsMicroservicesRabbitMQOOP

Frontend

Customer-facing web products, tuned with A/B tests and performance work.

ReactTypeScriptJavaScriptA/B testingWeb performance

Cloud and DevOps

From legacy servers to cloud pipelines that deploy many times a day.

AWSDockerKubernetesJenkinsSpinnakerCI/CDGitGitHub

Data and search

Relational at the core, with document stores, caches and search where they fit.

MySQLPostgresSQL ServerOracleDB2SQLiteMongoDBRedisElasticSearchOpenSearchApache SolrApache NiFi

Ways of working

Agile delivery, metrics-driven prioritization and the tooling around it.

AgileScrumKanbanJiraConfluencePostmanPyCharmDataGripCursorWindsurfShopify

Leadership

Building teams that get better every quarter, and platforms that let them.

Technical strategyArchitecture ownershipHiring and onboardingMentorship and coachingPerformance managementTechnical debt roadmaps

Education.

August 2024

Lambton College

Certified Artificial Intelligence Engineer

Toronto, Ontario

October 2013

Fundación Universitaria Los Libertadores

Bachelor of Computer Engineering

Bogotá, Colombia

Spanish, native. English, proficient.

Licenses and certifications.

Twenty-four credentials, grouped by what they sharpened. Each title opens the credential.

Machine learning and data

Fintech and agile delivery

Software architecture

Data, search and infrastructure

Quick facts.

The short answers, for anyone who arrived here with a question.

Who is Erwin Mongui?

Erwin Mongui is a software engineering leader based in Toronto, Ontario, Canada. He is Director of Technology and Software Engineering Manager at Cedar Planters, and a Certified Artificial Intelligence Engineer. He has 18+ years of experience building customer-facing platforms and leading the teams that run them.

What migration did Erwin Mongui lead at Fincaraiz?

As CTO of Fincaraiz.com.co he led a team of 30 through the move from a legacy ASP.NET monolith to Django and React microservices on Kubernetes, for a real estate portal with more than 5 million monthly visits. The migration ran with zero downtime, cut hosting and maintenance costs by 50%, and rebuilt search with natural language and map-aware matching.

Has Erwin Mongui published research?

Yes. He co-authored “Eliminating Randomness in Selecting International Passengers for RT-PCR” with Umid Ghimire, Dharmik Limbani and Sairam Kilari at Lambton College, for the STEM Fellowship Big Data and AI Challenge 2023. It was published in the STEM Fellowship Journal (doi:10.17975/sfj-2023-007).

Does Erwin Mongui speak Spanish?

Yes. Spanish is his native language and he works professionally in English. He built the first part of his career in Bogotá, Colombia, before moving to Toronto in 2022.

What does Erwin Mongui do at Cedar Planters?

Since January 2024 he has led Cedar Planters' engineering as Software Engineering Manager and Director of Technology in Oakville, Ontario. He owns the full software lifecycle from requirements through deployment, redesigned the data models behind analytical and operational workloads for 20% faster retrieval, and runs continuous improvement of the website through A/B testing and third-party integrations.

What is Erwin Mongui's background before Cedar Planters?

He was CTO and Software Engineering Team Lead at Fincaraiz.com.co in Bogotá (2019 to 2022), where he led a cloud migration that halved infrastructure costs and shipped 15+ REST APIs on a Python and Django microservice backend. Earlier he was Lead Software Engineer at Fincaraiz.com.co (2015 to 2018), Senior Software Engineer at Periferia IT consulting for Old Mutual and Editorial Legis (2013 to 2015), and Junior Software Engineer at Americas BPS for Hewlett Packard Enterprise (2010 to 2012).

What is Erwin Mongui's education?

Erwin Mongui holds a Bachelor of Computer Engineering from Fundación Universitaria Los Libertadores in Bogotá (2013) and is a Certified Artificial Intelligence Engineer from Lambton College in Toronto (2024). He is also a Scrum Master Certified (SMC) professional.

What technologies does Erwin Mongui work with?

Python, Django, C#, .NET, Node.js, React and TypeScript for building; AWS, Docker, Kubernetes, Jenkins and Spinnaker for shipping; PostgreSQL, MySQL, MongoDB, Redis, Elasticsearch, OpenSearch, Apache Solr and Apache NiFi for data and search; and TensorFlow, PyTorch, Pandas and NumPy for machine learning, with Cursor and Windsurf for AI-assisted development.

Let's talk.

Whether you're scaling a platform or building the team that runs it, I'd like to hear about it.

Email LinkedIn profile