Ignacio Ordovás Pascual, PhD
Senior Data Science Consultant · SDG Group · Madrid, Spain
Profile
I spent six years analysing X-ray and optical spectra of supermassive black holes before moving into industry. Today I’m a senior data science consultant at SDG Group, where I develop Data Science and AI solutions for companies and take on technical guidance and AI lead tasks. Among the projects I’ve been involved in, I’ve built RAG chatbots used in production, generative-AI assistants for logistics planning, computer-vision defect detection and statistical studies for clients in banking, life sciences, insurance, food and energy.
That scientific background still shapes how I work: I care about getting the question right, measuring things honestly and communicating results succinctly to non-technical audiences. I’m a curious, fast learner and I enjoy mentoring juniors (I’ve mentored graduates in data science and computer vision and directed a Bachelor’s thesis).
Experience
Senior Consultant, Data Science
Sep 2024 – Present
SDG Group · Madrid, Spain
Generative AI for pharma
- Developed several GenAI solutions for a pharmaceutical company, including a research-assistant chatbot.
- Built a content generator that creates assets such as emails, slide decks and medical documents grounded on approved material, together with a solution to review the generated content.
- Increased user satisfaction by around 40% with respect to the previous tools.
Leadership & internal initiatives
- Delivered technical guidance, development and team lead tasks.
- Supervised a Bachelor’s thesis.
- Published “Injecting company jargon for generative AI specific use cases” on SDG Group’s Medium publication.
Associate Data Scientist
May 2021 – Aug 2024
Cognizant Technology Solutions · Madrid, Spain
Generative AI & data science
- Developed a RAG-based chatbot over internal data, used in production at a banking company. Improved its performance, leading to the user base growing from 50 to 600 employees.
- Built a generative-AI chatbot connected to a graph database of supply-chain data, giving users automatic descriptive analytics and plots for logistics planning.
- Designed an NLP record-matching use case (sentence similarity, fuzzy string matching).
- Worked on optimisation use cases to find the most adequate supply-chain actions.
Statistics & science
- Analysed multidimensional survey data using causal inference, propensity score weighting and bootstrapping.
- Updated an AI model that helps clinical-study designers plan procedure schedules, incorporating feedback from users, product owner and experts.
Computer vision
- Built neural-network and ML solutions (autoencoders, clustering, among others) to detect defects in objects of interest.
- Created Azure Percept computer-vision models (an IoT device in preview) for object-recognition use cases.
Engineering & deployment
- Deployed AI demos to Azure Web Apps with Docker, Streamlit and Flask APIs; worked with the testing team on API endpoints to productionise solutions.
- Designed a GitHub CI/CD pipeline taking input data from S3 and deploying Flask APIs in Docker containers pushed to AWS ECR.
- Built MySQL connections in Python, generated reports and configuration files from CSV/XML with Jinja templates, and added logging and unit tests.
Mentoring
- Taught data science and computer vision concepts and applications to graduates.
Researcher (PhD)
Jan 2014 – Dec 2019
Institute of Physics of Cantabria (CSIC-UC) · Santander, Spain
Research
- Analysed a dataset of 258 Active Galactic Nuclei using proprietary observations and public spectra.
- Wrote and adapted Python code to detect emission lines and fit models across large spectral datasets.
- Planned numerical simulations to measure nuclear obscuration of AGN and its emission properties.
- Presented results in 10 talks and posters at conferences and seminars, including two invited seminars.
- Research stays at the Brera Astronomical Observatory (Milan, 2015) and the National Observatory of Athens (2016).
Teaching & outreach
- Taught astronomy lab sessions for the Physics degree at the University of Cantabria (35 hours, in Spanish and English).
- Coordinated outreach activities, including public observations, school talks and visits to IFCA.
Education
PhD in Science and Technology, cum laude
Jan 2014 – Dec 2019
Thesis A multifrequency view of Active Galactic Nuclei, supervised by Silvia Mateos and Francisco J. Carrera. I studied the relation between UV/optical extinction and X-ray absorption in type-1 AGN, finding that the two are not correlated and that both relate to the optical sub-classification. Summary · PDF · UCrea
Data Analytics Bootcamp
Oct 2020 – Dec 2020
Advanced Python, SQL and MongoDB, statistics, supervised and unsupervised ML with scikit-learn, neural networks with TensorFlow/Keras, Flask APIs, web scraping and NLP. Final project selected for the Hackshow. See the projects.
Master's Degree in Astrophysics
Sep 2012 – Jul 2013
Specialisation in Computation & Theory and Technology & Instrumentation. My first machine-learning project: a single-pass k-means that updates centroids on the fly, 20–40% faster than classic k-means on 20,000 galaxy spectra with equivalent classifications (Ordovás-Pascual & Sánchez Almeida 2014).
Bachelor's Degree in Physics
Sep 2005 – Sep 2012
Complutense University of Madrid & University of La Laguna
Started in Madrid and finished in the Canary Islands with a specialisation in astrophysics.
Skills
- Languages & core
- Machine learning
- Generative AI & NLP
- Computer vision
- Data & statistics
- Data stores
- Cloud & deployment
Publications
-
Extinction and AGN over host galaxy contrast effects on the optical spectroscopic classification of AGN
-
A multifrequency view of Active Galactic Nuclei
-
Machine learning in APOGEE. Unsupervised spectral classification with K-means
-
AGNs with discordant optical and X-ray classification are not a physical family: diverse origin in two AGNs
-
Searching for luminous absorbed sources in the WISE AGN catalogue
-
A fast version of the k-means classification algorithm for astronomical applications
Full list on NASA ADS.
Certifications & courses
- Databricks Certified Generative AI Engineer Associate (Sep 2026)
- Microsoft Certified: Azure Fundamentals (AZ-900) (Sep 2021)
- Azure Percept Bootcamp (Microsoft)
- The Complete GitHub Actions & Workflows Guide (Udemy)
- Complete TensorFlow 2 and Keras Deep Learning Bootcamp (Udemy)
- Learn BERT, the NLP algorithm by Google (Udemy)
- AWS Cloud Practitioner Essentials, AWS Cloud Technical Essentials and AWS Fundamentals: Addressing Security Risk (Coursera)
- Microsoft Azure DP-100, AZ-900, AI-900, DP-900 (Udemy)
- Machine Learning specialisation (Coursera)
- Introduction to TensorFlow for AI, ML and Deep Learning (Coursera)
- Hands-on Natural Language Processing with Python (Udemy)
- Python for Computer Vision with OpenCV and Deep Learning (Udemy)
- Business Etiquette 101: Social Skills (Udemy)
- Narrate the Science: scenic and oral narration for science communication (University of Cantabria)
- NEON Observing School 2015 (Asiago Observatory, Italy)