Petar Zečević, PhD
Big data and machine learning expert
I help companies turn large amounts of data into clear decisions and working AI systems. I have 25 years of experience in software development, and most of the last few years has been about AI. I hold a PhD in Computer Science from the University of Zagreb and wrote the book Spark in Action (Manning, 2016). I often speak at data science and big data conferences.
What I can help you with ...
Got a problem that is keeping you up at night? Pick the area that sounds closest, or let me pick one for you. Each one opens to show how I could help and which tools I could bring.
... building AI and language model applications
Wouldn't it be nice if your own documents could answer questions for you? I build AI assistants that read your files and data, look things up, and work through a task step by step. I start with the simplest design that works, try it on real questions, and keep an eye on cost and quality.
Tools in my toolbox
- LLMs
- LangChain
- LangGraph
- FastMCP
- Vertex AI
- Agent Builder
... machine learning and statistics on your data
Data often hides a good story, and finding it can sometimes bring real joy. I spot groups and patterns, build models that make predictions, and use Bayesian statistics to show how sure we can really be. My PhD was built using this, and I love the math behind the models as much as the tools.
Tools and methods in my toolbox
- Clustering
- Classification
- Bayesian statistics
- MCMC
- Keras
- TensorFlow
- R
- Python
... running machine learning models in production (MLOps)
A model that works on a laptop is a great start, but it has to survive the real world too. I help models move into daily life, with automatic training and testing, smooth deployment, and a clear plan for updates, all the while keeping an eye on quality.
Tools in my toolbox
- mlflow
- Kubernetes
- Ansible
- Cloud Run
... data pipelines for large volumes of data
I can help you design systems that collect, move, clean and process large amounts of data, in batches or as a live stream. I have built data lakes and replication systems for clients, and I helped keep them running smoothly as the data piles up.
Tools in my toolbox
- Spark
- Kafka
- Hadoop
- Hive
- NiFi
- Cloudera
- Hortonworks
- Fivetran (HVR CDC)
- Airflow
- MinIO
- Delta Lake
- Databricks
- HBase
- Cloud databases
- PostgreSQL
- Oracle
- DB2
... building and running systems on Google Cloud and AWS
Let’s get your systems into the cloud without any nasty surprises on the bill. I can plan and set up cloud systems that are easy to run and only use what they need, from data warehouses to machine learning platforms that grow and shrink with demand. I hold a Google Cloud Architect certificate, so I can plan the whole picture, not just single services.
Tools in my toolbox
- BigQuery
- Vertex AI
- Cloud Run
- Cloud Functions
- Compute Engine
- AWS S3
- AWS EC2
- AWS Redshift
... software for business systems and enterprise platforms
Business software is a bit like an old house. It works, but you have to be careful when you renovate. I build and connect the software companies rely on every day, from web apps and back-end services to links with older enterprise systems. Java has been my companion for many years, so I know how to add new rooms without bringing the roof down.
Tools in my toolbox
- Java
- Spring
- Hibernate
- Python
- Django
- JavaScript
- WebSphere
- MQ
- WebLogic
- Load balancers
- HTTP servers
Experience
From a junior developer in 2000 to leading AI projects today.
-
Self-employed
- Climatig: calculating climate risks and adding AI features
- AKD and Affinis/NSCP: ingesting and processing data with Spark, MinIO and related tools
-
Delivery Director for AI
- Managed a team of 7 people, including 2 PhDs
- Led projects in AI, machine learning and data engineering
-
Senior Principal Consultant
- Projects in AI, machine learning and data engineering
-
Senior Developer, Team Leader and CTO
- Big data projects for various clients: designing data lakes, data integration, replication and streaming
- IBM software consulting and implementation for various clients
- Java development for various clients
-
Junior Developer, then Senior Developer
- Built an internal product for industrial measurements, in Java and C++
Education, certificates and languages
Education
PhD in Computer Science, 2017 to 2023
Faculty of Electrical Engineering and Computing, University of Zagreb
Thesis: “Fast Distributed Cross-Matching of Big Astronomical Data and Parameter Estimation of Moving Point Source Model”.
I built Astronomy eXtensions for Spark (AXS), a system on top of Apache Spark with my own algorithm for fast matching of large tables. I also used Bayesian statistics and MCMC to create a new online method for estimating how a point source moves.
Check out my PhD Thesis here
Check out my AXS paper here
Master of Engineering in Computer Science, 1996 to 2002
Faculty of Electrical Engineering and Computing, University of Zagreb
Certificates
A selection of my certificates:
- Google Cloud Architect
- Databricks Data Engineer Associate and Professional
- Coursera data science courses
- Many IBM certificates
Languages
- Croatian
- English
- French
- Italian
- German
Let’s talk about your data
Tell me a little about your project, the problem you want to solve, and your timing. I read every message and reply within two working days.
- petar (dot) zecevic (at) this domain
- Zagreb, Croatia