Victor Hernandez
Senior Backend & Data Engineer · Python · Applied LLM systems
Granada, Spain · Remote or hybrid
Go to: Summary | Skills | Experience | Education | About me | PDF | Español
Summary
Backend and data engineer with 12 years of experience, the last 7 building a data-intensive SaaS platform: from a legacy monolith to microservices, large-scale data-collection pipelines and LLM features running in production. I own problems end to end, from understanding the need to designing, measuring and documenting the service. Open to remote or hybrid senior roles in backend, data or applied LLM engineering.
Skills
- Backend: Python, FastAPI, Django, SQLAlchemy, Pydantic; REST API design with OpenAPI and contract testing.
- Applied LLM engineering: Azure OpenAI in production, prompt design and output validation, evaluation harnesses, cost and throughput tuning.
- Data: PostgreSQL, MySQL, MongoDB, OpenSearch, CosmosDB, DuckDB, Parquet, pandas; ETL and scheduled pipelines; large-scale data collection from third-party APIs.
- Cloud and DevOps: Azure, Kubernetes, Helm, Docker, Terraform, GitLab CI; observability with OpenTelemetry and Grafana.
- Also: Flask, Stripe, SAS, Hadoop (Impala, Hive), Power BI, IBM WebSphere and Domino, Linux and Windows administration.
- Languages: Spanish native; English and Portuguese fluent.
Experience
Primetag (Portugal, remote) – Software Engineer (Backend & Data) – October 2019 to present
Marketing-analytics SaaS platform.
- LLM features (2024 to present). Built the platform's LLM analysis pipelines on Azure OpenAI from proof of concept to production, including automated alerts and the evaluation study that set the feature's cost limits. Core backend engineer on the platform's LLM assistant, scaling its analysis to millions of records.
- New services (2024 to present). Designed and built a new microservice end to end. One of the main contributors to the core campaign service and several supporting services.
- Data collection (2021 to 2024). Core engineer of the pipelines that collect social data at scale: Kubernetes workers and scheduled jobs, message queues, media processing, rate limiting against third-party providers. Expanded coverage to new markets on four continents.
- Legacy platform, reporting and reliability (2019 to present). Backend of the original Django application, including payments, third-party API upgrades and its migration to Kubernetes. Report generation and client-facing reporting APIs. Go-to engineer for client-facing incidents, from diagnosis to root-cause write-up.
Grupo Bancolombia (Colombia) – Software Developer, then Analyst – August 2015 to September 2019
Colombia's largest bank.
- Analytics and automation for business teams: data mining with SAS, Hadoop (Impala, Hive), SQL Server and DB2; process automation and reporting in Python; dashboards in Power BI and Qlik.
- Internal web applications and APIs with Python and Flask.
TSGroup (Colombia) – Systems Analyst – March 2014 to July 2015
IT services company.
- Linux and Windows infrastructure, automated provisioning for large-scale deployments, IBM WebSphere Portal and Domino support.
Autolarte (Colombia) – IT Support Assistant – July 2013 to January 2014
- End-user support, Zimbra mail and internal chat administration, and the rollout of an Asterisk phone system.
Education
- Bachelor's degree in Informatics Engineering, Universidad Católica del Norte (UCN), Colombia, 2020.
- Associate Degree in Applied Science in Computer Networking Technology, SENA, Colombia, 2013.
About me
From Medellín, Colombia, living in Granada, Spain. I run, travel, and tinker with a Raspberry Pi home server. You can also find me on LinkedIn.