Profilbild von Michele Dallachiesa Data Scientist, Data Engineer, Research Engineer aus Munich

Michele Dallachiesa

verfügbar

Letztes Update: 06.09.2022

Data Scientist, Data Engineer, Research Engineer

Abschluss: Ph.D. in computer science
Stunden-/Tagessatz: anzeigen
Sprachkenntnisse: englisch (verhandlungssicher) | italienisch (Muttersprache) | spanisch (gut)

Skills

  • Data Science and Machine Learning: machine learning and data mining algorithms, experience with Scikit-learn, TensorFlow, Seaborn, Pandas, NumPy, Joblib, Jupyter Lab.
  • Data Engineering: static and streaming data processing, experience with Cassandra, PostgreSQL, Parquet, HDFS, Hadoop, Spark, Apache Airflow, Celery, Fabric, Docker, Flask.
  • Location Intelligence: indoor and outdoor localization methods, cellular network data analytics, urban traffic modeling and optimization, experience with SUMO (Simulation of Urban MObility).
  • AWS services: EMR, S3, Lambda, CloudWatch.
  • Programming languages: Python, Java/Scala, C/C++.
  • Graph Mining: Neo4J, NetworkX.
  • Robotics: experience with ROS (Robot Operating System).

Projekthistorie

01/2019 - 12/2020
PC member
Technology (EDBT)

PC member. Serving in the Program Committee of the International Workshop on Big Mobility
Data Analytics (BMDA) in conjunction with the International Conference on Extending Database
Technology (EDBT) 2019, 2020. December 2018 - present

05/2017 - 06/2019
Lead Data Scientist, Senior Data Engineer
Minodes GmbH

Berlin, Germany. Delivering in-store and city-scale behavioral data analytics for use cases in retail.
May 2017 - June 2019

List of achievements:

* Coordinated the development of a GDPR-compliant analytics service for out-of-home advertising
based on the movements of people at city-scale, using the logs of the Telef´onica cellular
network, and running on AWS EMR clusters. The service has been successfully tested in the
Berlin and Munich city areas.
* Delivered an analytics platform based on a micro-service architecture (Airflow/Celery) to profile
people and their movements at city-scale using programmatic advertising data sources with
successful tests in the Berlin city area.
* Redesigned the existing WiFi analytics pipeline as a series of orchestrated Spark jobs, greatly
simplifying the complexity of the code base and increasing the quality of the localization
accuracy by 11% with hyperparameter optimization at no extra model complexity.
* Implemented a novel method to improve the accuracy of sequences of predictions by combining
classifiers with Hidden Markov Models (HMMs), improving the localization accuracy of the
WiFi analytics pipeline by 32%.
* As team lead, I had the privilege to hire and manage a talented team of four. I implemented
the best practices of agile development including daily standups, sprint planning, review,
retrospective, backlog refinement, regular 1:1 meetings, and adopted the OKRs framework.
Having joined the company right after its acquisition, I built bridges and coordinated the
collaborations between my team and the other groups within the Telef´onica family.

The company has ceased its operations due to a reorganization triggered by Telef´onica.

05/2014 - 03/2017
CTO and Founder
Skysense Inc.

List of achievements:

* Supervised the design, development, and commercialization of a variety of products, including
indoor and outdoor charge stations for drones, companion miniaturized battery chargers,
and remote-controlled hangars. These products required the development and integration
of complementary software services tailored for our specific needs, including the web management
dashboards, the robotics/IoT communication backbone, the vision-based precision
landing module, and the web-based mission planner for beyond-visual-line-of-sight (BVLOS)
operations.
* Managed our Intellectual Property (IP) assets, including the documentation of the hardware
production schemes and test processes, the documented source code, and the user manuals. I
have contributed to several inventions applied to the field of aerial robotics. This work also
resulted in a US patent.
* Management of operations, including handling of orders, supply chain, production, testing,
logistics, worldwide deliveries, technical support.
* Drafting and reviewing of NDAs, employment contracts, consulting contracts, collaboration
agreements, and patent applications.
* Delivered project proposals including executive summary, project overview, technical requirements,
list of assets, milestones, financial terms, and payment terms.
* In this position, I hired and managed a group of exceptional employees, contractors, and partner
companies involved in research, development, production, operations, and IP protection.
Headcount of five with spikes of up to ten units. With clear documented processes and responsibilities,
I adopted the best practices in agile project management in an international and
distributed team setting.
* My contributions led to contracts with more than thirty international clients, including CIA,
NASA, NIH, Google X, Parrot, and Stanford University.

The journey ended due to diverging opinions on integrity and transparency.

10/2011 - 06/2013
CTO and Founder
GraphInsight

to explore large networks with millions of nodes and connections. More than 600 users from the
academy and industry. October 2011 - June 2013

08/2012 - 11/2012
Research Intern
IBM Watson Research Center

NY, USA. Investigated collective classification algorithms for uncertain network graphs. Supervised
by Dr. Charu Aggarwal. August 2012 - November 2012

02/2012 - 04/2012
Research Associate
Qatar Computing Research Institute at Qatar Foundation

Designed a general framework to support user-defined data cleaning semantics on large datasets.
Supervised by Prof. Ihab F. Ilyas. February 2012 - April 2012

09/2011 - 12/2011
Research Intern
IBM Watson Research Center

NY, USA. Introduced methods for uncertainty management in the System S data stream processing
service. Supervised by Dr. Kun-Lung Wu. September 2011 - December 2011

07/2008 - 09/2010
Software Engineer and Founder
Reactive Search Srl

analytics tool to investigate the internal network dynamics of modern SAT solvers in collaboration
with Microsoft Research Cambridge. July 2008 - September 2010

09/2008 - 10/2009
Research Fellow
Machine Learning and Intelligent OptimizatioN (LION) Group, University of Trento

Trento, Italy. Implemented a distributed reputation system as part of the BIOlogically inspired
NETwork and Services (BIONETS) European project. September 2008 - October 2009

10/2006 - 01/2007
Software Engineering Intern
IT Security Department, Lutech SpA

and developed a C/C++ VoIP tapping module for intelligence operations to recognize, reconstruct
and record H.323, SIP, SCCP, and RTP voice data streams. October 2006 - January 2007

Reisebereitschaft

Weltweit verfügbar
I am based in Munich, available to travel.
Profilbild von Michele Dallachiesa Data Scientist, Data Engineer, Research Engineer aus Munich Data Scientist, Data Engineer, Research Engineer
Registrieren