I'm an HR professional with a strong technical profile and 6+ years of experience at companies such as Amazon and Olin Group. I built an HR department from the ground up. Since then I've specialised in automating processes with Python, analysing data with SQL and presenting insights with Power BI dashboards. I use Claude and ChatGPT every day to design, build and document these automations faster and with higher quality.
📍 Madrid, Spain
🎯 Looking to join teams that use AI to automate operations.
| Process automation | I map repetitive workflows end to end and replace manual steps with reliable Python scripts, improving operational efficiency and scalability. |
| Data analysis & reporting | HR dashboards and KPI reporting with Power BI + MySQL that give management data-driven insights. |
| Process design at scale | Policies, processes and governance frameworks built from scratch, including integrating several companies after acquisitions. |
| AI-augmented delivery | LLMs are part of my everyday toolkit: from prompt to working, documented solution. |
| Business + tech bridge | I understand the people and processes behind the data and translate business needs into technical solutions. |
| Data quality & privacy | Experience auditing and maintaining databases, and handling sensitive personal data with GDPR in mind. |
AI is a core part of how I work:
- Claude
- Designing automation scripts and reviewing code.
- Refactoring and writing technical documentation.
- Analysing long documents and reasoning through complex business rules.
- ChatGPT
- Rapid prototyping and SQL query generation.
- Data exploration and analysis of files (Excel/CSV).
- Drafting reports, and building custom assistants (GPTs) for repetitive tasks.
- Prompt engineering
- Structured prompts to extract, classify and transform information from unstructured text (forms, surveys, documents).
- Human in the loop
- I validate every AI output against business rules and real data before it reaches production.
AI does not replace understanding the process; it multiplies what you can build once you understand it.
- Automation & programming: Python (pandas, openpyxl, numpy, argparse)
- Data & BI: SQL (MySQL), Power BI, advanced Excel, KPI design
- AI: Claude, ChatGPT, prompt engineering
- HR systems: Factorial, Sage, Microsoft Forms
- Business skills: HR analytics, HR transformation, M&A integration, process design, stakeholder management
Problem: HRIS exports arrive as messy Excel files: title rows, totals, inconsistent headers, accents and invalid dates. They can't be loaded into a database as they are.
Solution: a Python data-cleaning pipeline that:
- detects the header row and normalises column names;
- validates the required fields and applies business rules;
- standardises the dates;
- outputs a clean, SQL-ready CSV that feeds the Power BI dashboards.
Impact: a manual, error-prone cleanup becomes a repeatable, auditable step in the reporting workflow.
Python · pandas · openpyxl · numpy
Problem: creating one personalised data-protection consent document per new employee was done by hand.
Solution: a CLI tool that reads Microsoft Forms responses (exported to Excel), fills a Word template and generates one print-ready PDF per person. It also:
- filters responses by date;
- validates the input;
- names files without ever overwriting a duplicate;
- keeps personal data out of the console output.
Impact: an end-to-end onboarding document automation; a batch takes minutes instead of hours, with fewer copy-paste errors.
Python · pandas · argparse · document & PDF generation
What it is: a reusable library of analytical SQL queries over monthly workforce snapshots. It covers headcount evolution, hiring, tenure bands, age distribution, contract types and working-hours analysis.
Why it matters: the queries behind the KPIs; recurring business questions answered in a reusable way.
SQL (MySQL) · UNION ALL · aggregations · date functions
Understand the process → Clean & validate data → Automate (Python + AI) → Visualise (Power BI) → Document → Iterate
- Process first: I automate what I understand; automating a broken process only makes it fail faster.
- AI as a co-pilot: Claude and ChatGPT speed up design, coding and documentation, and I review everything they produce.
- Data you can trust: early validation, clear errors, safe defaults.
- Insights people can use: dashboards and KPIs that drive decisions, not just raw tables.
- Document for the next person: every project ships with a README covering its flow, rules and limitations.
- Integrating LLM APIs into automation scripts (classifying and extracting information from documents)
- Natural-language querying over SQL data
- Low-code automation for end-to-end business workflows
- 💼 LinkedIn: linkedin.com/in/nuria-cg
- ✉️ Email:
nuriacerdangarcia@gmail.com
Open to conversations about process automation, data analytics and AI-driven operations.