Max Flander, Developer in San Francisco, CA, United States
Max is available for hire
Hire Max

Max Flander

Verified Expert  in Engineering

Software Developer

San Francisco, CA, United States
Toptal Member Since
March 31, 2016

Max has been working as a data scientist since 2013, backed by a Ph.D. in mathematics. His areas of expertise include data engineering, predictive modeling, recommender systems, information retrieval, and crowdsourced data curation. With exceptional communication skills and his knack for cutting to the heart of the discussion, Max has worked in Agile teams of varying sizes, and he is known for running highly effective stand-ups, planning, and team retrospectives.



Preferred Environment

Amazon Web Services (AWS), Docker, Git, Linux

The most amazing...

...thing I've built is a 20GB per day AWS S3 JSON to HDFS/Impala ETL pipeline, using Hive and Hadoop Streaming.

Work Experience

Data Scientist

2014 - PRESENT
SEEK, Ltd.
  • Performed data analysis and modeling, using Python, Pandas, and scikit-learn.
  • Worked on a team that was building a Haskell ETL pipeline.
  • Implemented real-time data visualization using D3, Google Maps, and Tableau.
  • Built a 20GB per day AWS S3 JSON to HDFS/Impala ETL pipeline, using Hive and Hadoop Streaming.
  • Crowdsourced a data curation project, using CrowdFlower.
  • Built and deployed a batch-scored predictive model running on six million records per week on a SQL Server.
Technologies: Amazon Web Services (AWS), Spark, Hadoop, Python

Data Scientist

2013 - 2014
Predictive Match
  • Built a recommender system for a real estate website, using Python.
  • Developed a prioritization model for recommendation algorithms using RapidMiner.
  • Implemented data analysis and visualization, using a Python data science stack.
  • Performed clustering analysis of users, using scikit-learn.
  • Collaborated using tools such as Trello and Bitbucket.
Technologies: Amazon Web Services (AWS), RapidMiner, Python

eBay Sale Predictor in Python

A Python script that scrapes and prepares a dataset from eBay, trains a gradient-boosted tree-ensemble model using Scikit-learn, then predicts the probability of the listing ending in a sale.

I was inspired to build this toy because I was bored with sifting through unwanted ads on eBay. The model performed well—with a ROC AUC of 0.7 on the example I looked at—considering it is trained only on the ad title and price. All the dependencies were included in the Anaconda distribution of Python 3.4.


Python, SQL, Clojure, Python 3, Ruby, Haskell, JavaScript


Apache Spark, Hadoop, Mach-II, Ruby on Rails (RoR), Flask, Spark, Windows PowerShell, ClojureScript


Pandas, Requests, Scikit-learn, Matplotlib, TensorFlow, PyTorch


Amazon Elastic MapReduce (EMR), Apache, Git, Tableau, Emacs, AWS CloudFormation


Agile Software Development, Mobile Development, Continuous Delivery (CD), Functional Programming, Continuous Integration (CI), Unit Testing


Amazon Web Services (AWS), Docker, Linux, Unix, Kubernetes, Windows, Windows Server 2012, RapidMiner, Anaconda


Redshift, Amazon S3 (AWS S3), Microsoft SQL Server, Elasticsearch, PostgreSQL, MySQL, MongoDB

Industry Expertise



Web Scraping, Regular Expressions, Smart Contracts, ADK, Quality Assurance (QA), Deep Learning, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT), Integration Testing, Mathematics, eBay, Gradient Boosted Trees, Scraping

2009 - 2013

Ph.D. in Mathematics

University of Melburne - Melbourne, Australia

2003 - 2006

Bachelor of Science (Honors) Degree in Mathematics

University of Melburne - Melbourne, Australia