Freelance consulting for teams that need senior help with search, data or cloud, from the first workshop to a working proof of concept.
Technical presales & solutions architecture
Who it's for
Software vendors, partners and startups that need a senior presales architect for a deal, a launch or a demo program, without a full‑time hire.
What you get
Discovery workshops that turn business goals into a solution design
Demo scripts and POCs built around your customers' business questions
RFP/RFI technical responses and C‑level presentation decks
Evidence
Senior Solutions Architect at Elastic, technical point of contact for accounts in Mexico City and Monterrey. Presales roles at Datadog, Verizon, Axway and Kaspersky. Repo: snowflake-intelligence (opens in a new tab), a demo script organized in acts.
Search & AI with Elasticsearch
Who it's for
Product and engineering teams that want to move from keyword search to semantic, vector or hybrid search, or try AI agents over their own data.
What you get
A search assessment comparing BM25, k‑NN and hybrid (RRF) on your own data
Index, mapping and embedding design, with precision measured on your test queries
A working proof of concept (Java/Spring Boot or API) your team can extend
Evidence
Repo Elastiflix (opens in a new tab): 81 % top‑1 precision on 32 plot queries over 6,899 movies (1024‑d). Current role at Elastic.
Data engineering & analytics
Who it's for
Companies and system integrators that need to design or review data pipelines and analytics on Google Cloud or Snowflake.
What you get
Architecture and design of end‑to‑end pipelines for analytics and AI/ML
Hands‑on workshops so your team can run it on its own
Evidence
Data Engineer at Google Cloud Consulting (2024). Senior Partner Solutions Engineer at Snowflake (2025–2026). SnowPro Core. Repo: il-giro-de-fito (opens in a new tab).
Cloud program management
Who it's for
Organizations running a cloud program with several teams or vendors that need an independent program manager to keep scope, risk and timelines visible.
What you get
A program plan that breaks big goals into executable workstreams
Risk, issue and action tracking with mitigation plans
Delivery assurance and status reporting across vendor teams
Evidence
Cloud Program Manager for Google Cloud Professional Services (2022–2024). Project Management diploma (ITAM). Professional Cloud Architect. Repo: datadog-gcp-integration (opens in a new tab).
/en/#projects
Projects
Elastiflix
Problem
Teams hear "semantic" and "hybrid" search but rarely see why a result ranks where it does.
Solution
A movie search app that puts full‑text (BM25), vector (k‑NN) and hybrid (RRF) results side by side and shows where each score comes from. The earlier version compares six modes, including ELSER, E5 and cross‑encoder reranking. Bilingual UI.
Stack
Java 25
Spring Boot 4.1
Thymeleaf
Elasticsearch Serverless
jina-embeddings-v5
Checkstyle/PMD/ArchUnit
JaCoCo
Result
6,899‑movie catalog. Top‑1 precision on 32 plot queries: 69 % (256‑d), 78 % (512‑d), 81 % (1024‑d). 512‑d gets close to 1024‑d at half the size. Previous version: 98.6 % instruction / 92.4 % branch test coverage.
Inspired by Elastic's original Elastiflix demo; built for academic and demo purposes.
Gotham News & Media Browser
Problem
A newsroom archive with text, images and media is hard to explore with keyword search alone.
Solution
A prototype media browser with semantic, hybrid, vector and multimodal search, built on Elastic Cloud Serverless with Google Cloud Storage and no relational database. Ships with architecture docs, UI mockups and a synthetic‑data generator.
Stack
Java 25
Spring Boot 4.1 (Maven multi-module)
Elasticsearch
ImageBind (1024-d)
Google Cloud Storage
Docker Compose
Ollama
Result
42 of 42 planned implementation tasks completed (phases P0–P10).
Connecting Datadog to Google Cloud means enabling many APIs, creating resources and routing logs by hand, and log forwarding is not configured by default.
Solution
Bash scripts that enable the required APIs, create a dedicated observability project, and set up a log router sink to forward logs. An extra folder covers Cloud Cost Management.
Stack
Bash
gcloud CLI
Cloud Logging
Datadog
Result
Public and reusable. It is his most‑starred pinned repo (5 stars as of Nov 2026).
Partners need a repeatable way to show what Snowflake's AI features do for a business, not just a feature list.
Solution
A demo script for a fictional pizzeria, organized in six timed acts: sentiment and summaries with Cortex LLM functions, natural‑language questions with Cortex Analyst, and sales forecasting and anomaly detection. It includes talking points and common Q&A.
Stack
Snowflake
Cortex Analyst
Cortex LLM functions
ML functions
SQL
semantic model (YAML)
Result
Presales asset created during his Senior Partner Solutions Engineer role. It is a demo with a fictional dataset, so it reports no results.
Snowflake's platform objects are easier to grasp in one coherent story than in scattered docs.
Solution
A notebook that builds a fictional Tuscan restaurant end to end in 12 sections: table and view types, Iceberg tables, stages and COPY INTO, Java and Python stored procedures, UDFs, streams and tasks, dynamic tables and RBAC.
Stack
Snowflake
SQL
Python
Java
Jupyter / Snowflake Notebooks
Apache Iceberg
Result
Presales and enablement asset, with no results reported.
Laptops on a private LAN and a tablet elsewhere need one Client VPN endpoint, without sending internet traffic through AWS or leaving persistent access keys on the devices.
Solution
Terraform reference architecture for an AWS Client VPN hub with split tunnel, private Route 53 DNS and IAM Roles Anywhere.
Stack
AWS
Terraform
Client VPN
Route 53
IAM Roles Anywhere
Result
Public and reusable lab reference. Example names, addresses and the account id are fictional, so it reports no production results.
Technical point of contact for accounts, supporting account executives in Mexico City and Monterrey. I take deals from discovery to the technical win.
I demo Agent Builder: chat‑style agents on commercial or local LLMs that search across all ingested data. I also enable partners and speak at meetups and conferences.
Senior Partner Solutions Engineer · Snowflake
–
Led technical strategy with system integrators to speed up adoption of their Snowflake-based offerings, and co‑built joint go‑to‑market solutions.
Enterprise Sales Engineer · Datadog
–
Ran demos and technical evaluations (POVs) for customers in Colombia, Peru, Ecuador, Central America and the Caribbean, and fed customer and competitive input back to product.
Google Cloud
–
Data Engineer, Google Cloud Consulting
–
Designed data migrations and end‑to‑end pipelines for analytics and AI/ML, and helped build scalable, highly available data‑processing systems.
Cloud Program Manager
–
Ran multi‑team Professional Services programs, including planning, delivery assurance, and risk and mitigation, plus partner delivery quality.
Infrastructure Cloud Consultant
–
Designed and deployed distributed cloud infrastructure as a trusted advisor to customer leads and executives.
Sales Engineer · Verizon
–
Big data and IoT presales focused on telecom carriers in LATAM. Worked on POCs, RFPs and C‑level presentations.
Presales Consultant / Solutions Architect · Axway
–
First technical contact for customers and partners, designing software solutions for complex scenarios.
TSE / Technical Account Manager · Kaspersky
–
Presales support, plus the bridge between support and premium customers, with SLA KPIs.
Senior Technical Support Engineer · Fortinet
–
Advanced support for firewall, routing, switching and VPN across LATAM, mainly Brazil.
Earlier · 2006 – 2012:Operations and IT roles at IBM and Acision (Brno, Czechia) and Jinny Software (Panama City), supporting messaging middleware for mobile operators and enterprise servers.
/en/#certifications
Certifications
SnowPro Core
Data & Information Governance (ITAM)
Google Cloud Digital Leader
Associate Cloud Engineer
Professional Cloud Architect
Professional Cloud Developer
Stack
Search & AI
Elasticsearch
Elastic Cloud Serverless
Agent Builder
ELSER
E5
k-NN
RRF
Embeddings
Data
Snowflake
Cortex AI
Apache Iceberg
Dynamic Tables
SQL
Python
Jupyter
Cloud & Observability
Google Cloud
Cloud Storage
Cloud Logging
gcloud
Datadog
Docker
Development
Java 25
Spring Boot
Thymeleaf
Maven
Bash
/en/#education
Education
MBA
Instituto Tecnológico Autónomo de México (ITAM)
2013 – 2015
Diploma in Project Management
ITAM
2016 – 2017
Systems Engineering degree (Ingeniero de Sistemas)
Universidad del Norte (Barranquilla, Colombia)
1998 – 2004
French, B2 (CCRE)
Université de Nice-Sophia Antipolis
2010 – 2011
/en/#about
About
Barranquilla
Nice
Brno
Panama City
Mexico City
I studied in Barranquilla and Nice, worked in Brno and Panama City, and now call Mexico City home. I work in English and Spanish every day, and I get by professionally in French and Portuguese.
Away from the keyboard you'll find me at the gym, with a comic or a movie, or playing guitar. I'm an aspiring chef who loves trying new food, and the beach is my favorite place on earth. On quieter nights it's a good series, or more reading about cloud and software, because that habit never really switches off.
/en/#contact
Let's talk
Have a search, data or cloud project in mind, or want me to speak at your event? Tell me a bit about it and I'll get back to you.