working experience





mobility, numeric
e-commerce, advertising

Dr Giovanni Marelli


event collection
overview of experiences

computational biophysics

reasearcher, teaching - Göttingen/Padova

research assistant 08-12

viral fusion

fusion pathway and activation probability for viral fusion

bilayer membranes

different topologies of bilayer membranes

fusion pathway

fusion pathway

inclusion objects

event collection
amphiphile inclusions in lipid membranes

phase transition

event collection
topology changes under lipid configuration changes

numeric calculation

event collection
numeric calculation for differential equations


event collection
teaching, exercises, mentoring

signal theory

C++ native software for signal analysis and plotting

studies and programming

simulation and research in statistical physics

solid state

research in solid state, defect diffusion in silicon

guide - teacher

teacher - guide: museo della fisica, sperimentando

enterprise planning

ERP/CRM developer and integrator - Padova/Brescia

data consultant 13-14

offgrid electrification

feasibility study for offgrid electrification in Tanzania

databases - forms

db ingestion/reporting: milk factory, hospital, power plant

ERP CRM inventory

digitalisation of small firms, cloud solutions

webapp - portals

bringing ideas to user interfaces


client based database manager - offline functions

electric violins

double funds to create viudì: electric violins

creative coding - arduino

openCV, openGL, openFrameworks, processing, arduino

d3 data visualization

data visualization and dashboards with d3.js

targeted e-commerce

A/B tests, conversion uplift


Data scientist 14-15

product owner

product owner over product lifecycle


personalizing the display - conversion uplift

booking funnel

personalize user history

a/b testing

monitoring a/b tests

right lure, right fish

how to distinguish customer classes

onsite retargeting

simple paths prevent leakage, keep the user

text mining

NLP, listen to your customers

customer feedback

Sentiment, promoter score, personas


implement and connect trackers

targeted advertising

look alike, conversion uplift


Sr Data Scientist 16-17

audience affinity

from browsing behaviour understand audience affinity


understand regional differences

time series

time series forecast, exogen features


event collection
forecast on signal components

socio demo

socio demo, post evaluations

data pricing

data pricing per format, device, brand...

data visualization

data exploration with d3.js


content classification

time series dashboard

dashboard with live time series

audience offer

presenting a large number of targets


organize and connect trackers, sync data and ids

page labelling

reliable content labelling

mobility - transportation

movement patterns, commercial capture rate


ML Expert 17-19

event collection

event collection
we collect an handover between cells per device (D1-Netz)

events to movements

event collection
we groups events into activities and trips

activity probability

event collection
density probability of activities between cells

daily patterns

event collection
people have different patterns every day

tourist patterns

event collection
split nationality, origin, counting activities

age, gender, device types

event collection
distribution of age classes per zip code

movements across the country

event collection
how people move during a day

most frequent origin and destination

event collection
important for local transportation companies


event collection
we work on an efficient routing and infrastructure

local network

event collection
special modules consider city centers

subway module

event collection
distinguish the mean of transportation labelling cells (training data)

subway passangers

event collection
collecting information about commuters

intensity of activities

event collection
activity intensity: breaks of motorway drivers

collect important features

event collection
we select the most relevant features

training set

event collection
geographical information to train system


event collection
we train our models to predict important features

compare external sources

event collection
public information from different sources

capture rate

event collection
calculate the coverage of a commercial activity

whitespot analysis

event collection
customer potential on regions not covered within a isochrone

time dependent data

event collection
weather/time dependent data

geo static information

event collection
predict absolute values

learning process

event collection
scoring, cross validation, classification

evaluate data and time series

event collection
series clustering, distribution, class stability


demand prediction, fleet optimization


AI lead 19-20

demand - offer areas

event collection
activation, revenues, app opening, errors

location intelligence

event collection
spot the relevant geo data for predictions

motion patterns

event collection
start, end, route, deviations, graphs

fleet dispatch

event collection
optimization engine for fleet dispatch

graph routing

event collection
from street network

ai - reinforcement learning

event collection
reinforcement learning to tune the simulation

infrastructure design

event collection
microservice based, parallel, asynchronous


event collection
interaction frontend, ui interaction with the solution

documentation - libraries

event collection
documentation libraries on github

Call center optimization

Metric/infra design, client stakeholding


Director of advanced analytics 20-22

performance key factors

event collection
driving features, feature dependency

metric design

event collection
Metric as a tradeoff of revenue and delivery capability

customer survivability

event collection
Churn factors, customer journey, survival rate

performance monitoring

event collection
monitoring KPIs, performances, drivers

feature importance

event collectiondrivers for conversion rate, uplift, maturation

maturation periodicity

event collection
check drivers week by week, monitor feature quality

data room

event collection
custom data visualization with js libraries

agent compensation

event collection
Simulation around the most effective reward

data pipe

event collection
infrastructure design, data pipes

strategic data consultant

Infra and language models - Berlin

freelance 22-23

designing and deployng infra

event collection
Analytical platform on prem

data pipelines

event collection
creating ETL jobs and schedule them

email outreach corrector

event collection
language model for email outreach

geographical anonymizator

event collection
loads a file and parse sensitive data before dumping to a db

logistics mini hubs

event collection
spots the best location for mini hubs (temporary parcel storages)