Francisco Sánchez Noguera

Tech Lead · AI & Data Platform

BASF

May 2026 — Present

Promoted in May 2026 to Tech Lead of BASF's AI & Data Platform team, leading 13 developers on a platform that hosts 800+ use cases and their infrastructure. The team provides governance for BASF's AI use cases, manages LLM consumption, and enables AI-related use cases. Also one of the main points of contact for the Databricks account (600+ workspaces, Unity Catalog, access to external data, and data governance).

• Lead 13 developers on a platform that hosts 800+ use cases and their infrastructure. Plan the technical work with the team and encourage teammates to build ambitious solutions
• One of the main points of contact for Databricks at BASF and for the company-wide account: 600+ workspaces, their Unity Catalog setup, access to external data, and data governance
• With the team, provide governance for BASF's AI use cases and manage LLM consumption. The team also enables AI-related use cases
• Proposed moving the platform's infrastructure as code from CDKTF to plain HCL Terraform. Led the first draft, then kept working on it with the developers
• Work closely with the architects on the platform design

Senior Software Engineer

BASF

January 2025 — May 2026

Joined BASF as an internal hire to continue platform engineering work, contributing to architecture decisions for a new data platform built on Azure, AKS, Databricks, and infrastructure as code (CDKTF and Terraform).

• Largest contributor to the Databricks and Unity Catalog infrastructure constructs on the CDKTF-based platform
• Designed a state migration process that used scripts to extract IDs and import resources without affecting users
• Resolved a production incident in Unity Catalog within 1 hour by auditing tables and reverting configuration changes
• Participated in the ADRs for the role-based permissions model and implemented role-based access in the portal and infrastructure
• Collaborated with architects on network topology design for the new platform
• Worked on logs and metrics ingestion across multiple clusters with Grafana OSS, configuring Alloy collectors, receivers, and scrape pipelines

Senior Data Engineer

Capitole Consulting (Client: BASF)

March 2023 — December 2024

Consultant data engineer working on two engagements at BASF: first a cosmetics analytics data platform, then lead developer for the Azure data platform provisioning infrastructure (300+ workspaces).

• Delivered a 12-month cosmetics analytics release in under 6 months using Medallion architecture with Delta Lake processing
• Led the refactoring of the platform provisioning codebase (Terraform + Python), raising the pipeline success rate from 20% to 90%
• Implemented a manifest change evaluator that reduced pipeline execution time from 1.5 hours to 30 minutes
• Cut the provisioning of a single platform from 4 hours to 20 minutes by orchestrating GitHub Actions workflows (Terraform for infrastructure, Python for configuration)
• Built a REST API with FastAPI and Container App Jobs for federated Databricks Account operations
• Implemented the Unity Catalog rollout across 300+ Databricks workspaces with a 1% incident rate
• Led a team of 5 engineers, increasing test coverage from 10% to 85% through code reviews, pair programming, and Python best practices
• Conducted ~30 technical interviews for Senior Data Engineer positions, resulting in 6 hires
• Managed the Data Community at Capitole: biweekly newsletter, meetups, and tech talks reaching 100+ data engineers

Big Data Developer

Kenmei Technologies

January 2022 — March 2023

Promoted to the Big Data department, processing massive telecom datasets with Scala and PySpark on Google Dataproc and Azure Databricks.

• Refactored and optimized a PySpark parser (Azure Blob Storage to Delta Lake), cutting job time from 20 hours to 4 hours (-80%)
• Reduced the client's cluster costs from €400/day to €80/day through caching and partition tuning, saving 100k+ EUR a year
• Productionized the Uplink Interference detection algorithm using distributed PySpark, reducing execution from 60 minutes (single node) to 6 minutes (-90%)
• Built a geospatial classification product with Scala and Sedona, categorizing tiles as indoor/outdoor across entire countries
• Maintained and extended a complex real-time trace processing system in Scala/Spark with a very small contributor base, and contributed to its 5G implementation

Junior Innovation Engineer

Kenmei Technologies

June 2019 — January 2022

R&D engineer working directly with the CTO on individual projects for telecom clients, building high-performance Python services and signal processing algorithms.

• Built a real-time drone geolocation service achieving 100+ geolocations/second (16× the client target) using NumPy vectorization and async MQTT
• Developed a high-throughput TCP trace server processing ~550 Mb/s with Cython and Numba; throughput was limited by the network interface rather than the CPU
• Created distributed coverage maps with Dask spatial partitioning, reducing country-level processing from 4 hours to 35 minutes (-85%)
• Designed an Uplink Interference detection algorithm using FFT and RSRQ/SINR metrics for PIM pattern identification, later productionized in the Big Data team
• Received recognition from a major telecom client's CTO for project delivery quality

CubeSats Concurrent Engineering Workshop

European Space Agency (ESA)

January 2019 — June 2019

University project selected for the ESA Concurrent Engineering Workshop. Worked in the "CubeTester" team to design a CubeSat concept for testing electronic devices in orbit.