Learn Data Skills
Beta
Piero Antonio Posocco

Piero Antonio Posocco

Course Trainer

Railsr | London

Technologies

My Portfolio Highlights

My New Course

Introduction to Python

My New Track

Data Scientist

Champion of data ethics, navigating the fine line between insights and privacy.

My Work

Take a look at my latest work.

course

Python Data Science Toolbox (Part 1)

course

Introduction to Python

course

Intermediate Python

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

My Work Experience

Where I've interned and worked during my career.

QA Ltd | Feb 2024 - Present

Project Lecturer — 3 month contract

Developing Content for the Advanced Data Science Course

Railsr | Apr 2023 - Oct 2023

Analytics Engineer

Reporting to the Data Engineering Manager I worked on creating Data Pipelines (DBT) and Dashboards (Looker) for the Risk and Compliance Team.
Show More

Railsr | Nov 2021 - Mar 2023

Lead Decision Scientist (Credit Risk)

Global Head of Risk as the main stakeholder I built Railsr first credit card risk model (Python) using TransUnion retros and optimised it for the Near Prime space, created proprietary scores and defined rules (pass/decline and line assignment), completed documentation and obtained the necessary Credit Risk Committe’s sign-off. I also created a proprietary Identity Fraud Risk Score based on Socure data that increased the acceptance in the first automated stage by 30%. - FinCrime: Working with the FinCrime, Transaction Monitoring team and the Global Head of Compliance as the main stakeholder I have been involved in drafting the integration of Featurespace with Railsr systems. I am also responsible for setting up the decision engine, preparing data ingestion and tuning the ML models. I’ve been supporting the MLROs with Risk analyses of customers and endusers, drafting risk scorecards and integrating the model in the onboarding process. - Regulatorors: I have been supporting the submission of reports for several European regulators (BoL, FCA, BNB, ...) with ad-hoc data pulls. That includes data sourcing, dashboarding and consolidating datasets across different domains (banking, cards, enduser PII, fraud, SAR, TM, AML, ...) and the automation of these pipelines (DBT). - Customer: With the Chief Commercial Officer as the main stakeholder I managed a Snr Data Scientist and a Salesforce Implementation Consultant working on Railsr lead conversion model with the aim of identifying and easing pain points in the Sales and Integration funnel using Salerforce data. The model and its most relevant features were then ported to Einstein Analytics for an easy(-er) integration within Salesforce.

General Assembly UK & Europe | Jul 2021 - Oct 2021

Data Science Immersive - Lead Instructor (DSI 19) — 3 month contract

Teaching, advising, and content development for the 12-week immersive GA bootcamp. Collaborating alongside permanent staff in order to best meet the needs and learning styles of the students. Inspire, motivate and mentor students to persevere through the challenges of learning complex subjects and eventually land a job. All the 5 students I was responsible for were offered a new position internally or a new job within 6 months from the end of the course.

FIBR | Jul 2020 - Jun 2021

Lead Data Scientist (Credit Risk) — 12 month contract

Reporting directly to the Head of Risk / Chief Risk Officer I covered the following areas: - Credit Risk Modelling: Defining bureau retros (Experian) and building the first commercial credit risk model (Python) for FIBR flexible funding line product. Defining data structures and ETL processes to deploy such models (AWS Sagemaker) and monitor (Metabase) their performance over time. Developed and deployed a second generation of the same model to include Company Directors’ information, which boosted the first model’s Gini by 20%. - Affordability Checks: Based on Open Banking, Bureaus and Companies House Data (API) I built a series of checks to inform Underwriters about the prospective clients free cash flow and built an affordability model that would produce an indicative offer later to be validated by more in-depth manual checks. That included future debt repayments, utilities, diversification of suppliers / customers, etc… - Analytics: Set up several dashboards in Metabase to track application quality, funnels, underwriting processes and overall efficiency of operations. - Product: With the Chief Product Officer as the main Stakeholder I managed a Snr Data Scientist and a Jnr Data Scientist supporting the Product team on their quest for improving the onboarding experience, A/B testing, data science enabled solutions (instant offers, conversion, etc…) to enable an easy integration with credit products marketplaces.

Liberis | Jun 2019 - May 2020

Head of Decision Science (Credit Risk)

manager of a 3 people team (Jnr Data Scientist, Data Scientist, Senior Risk Analyst): -Leading the Credit Risk Scoring activities: Laid out a programme aimed at improving the existing credit risk scorecards (Equifax). Based on an Agile approach and preconfigured sprints, the scorecards were rewritten in a modular fashion, monitored against latest write-off data, then retrained and deployed (MS Azure). Combining all scorecards into one resulted in increasing the overall performance (Gini) by 25%. Boosting Automation: Leading the team on a quest for the simplification of heuristics, raising funding thresholds, redefining scorecards boundaries. This led to an overall improvement of 10% of the automation related KPIs. - Making Open Banking a success: With the Global Head of Sales as the main stakeholder I was directly involved in popularising the importance and benefits of having customers go through the process of sharing their bank account information with Liberis. In addition to creating an invaluable proprietary dataset for customer segmentation, competitor analysis and transactions forecasting, we were able to cut the Sales TaT by 1 full working day. - Classification of transactions: The team built a series of dashboards based on transactions data (Open Banking) aimed at supporting the decisions of the underwriters. We productionised a couple of models (NLP, string similarities, Regex) which labelled inbound transactions by source (card volumes, wire transfers, cash deposits, etc...) and outbound payments (HMRC payments, wages, etc...). - Data Science for Marketing: Lead scoring model (XGB) based on partners and proprietary data, estimated to increase conversion by 30% compared to randomly selecting customers from a list.

Times Higher Education | Nov 2017 - Jun 2019

Data Scientist / Product Manager

- Developed and managed the entire data pipeline for a new product aimed specifically at the European market: identifying new data sources, driving the data collection (surveys, direct submissions, third-party), producing data validation tools, formulating the methodology, preparing robustness analyses, testing the market. - Working with the Engineering team to integrate Data Science approaches into existing and new products like: (a) dashboard reporting keys stats of a undergoing survey to highlight issues with sample size and balance, (b) recommender system for selecting customer’s peers, (c) sorting routine based on pairwise comparisons powered by Bayesian statistics. - Supporting the Editorial team with content and insight. Helping the Sales team to integrate Customer Analytics into their workflow. Presenting the product to senior university management to elicit participation in the data collection phase. Training the Customer Support team and answering customers’ technical queries on product features.

Imperial College London | Apr 2012 - Oct 2017

University Lecturer in Physics

- Worked on novel technologies for Medical Physics, Cancer Therapy with Charged Particles. Attracted funding in excess of £150k split between several small projects. Presented my work at international conferences & workshops. Lectured on MRI and Nuclear Diagnostics and tutored on many UG and PG subjects. In charge of the organisation of the High Energy Physics Outreach. Supervised 16 summers students, 1 MRes and 1 PhD student. - Scoping and leading projects in a highly multidisciplinary field with experts from different countries. Mentored and supported tutees in different stage of their career. Represented the interested of my group at various committees.

University of Exeter | Jun 2017 - Aug 2017

Data Scientist (Consultant)

Worked as a consultant for the ASI in the UoE Admissions Team on applying predictive analytics to the pool of university applicants with the aim of maximising offer acceptance and diversity, by means of (a) predicting students’ performance throughout their studies and (b) identifying those students from disadvantaged backgrounds who have unrealised potential to succeed at university. Have a look at the video below:

ASI Data Science | May 2017 - Jul 2017

Data Science Fellow

- A competitive eight week fellowship designed to be a bridge for academics transitioning into commercial data science - The programme provided training in the fundamentals of data science and engineering, including statistics, machine learning, databases, distributed computing and business skills

CERN | Oct 2009 - Mar 2012

Senior Fellow

RESEARCH: - SPL beam dynamics studies. (1) lattices and their sensitivity to the quadrupoles misalignment. (2) H- stripping in quadrupole magnetic field as function of the lattice, beam misalignment and distributions. (3) effect of the RF power distribution scheme on the longitudinal beam dynamics. - Linac4 LEBT commissioning. Linac4 LEBT beam characterisation measurements and in the development of a novel iterative method for emittance reconstruction in presence of space charge (code: Fortran). - ASACUSA re-commissioning. Collaborating with the anti-proton experimental line ASACUSA at the Antiproton Decelerator with the aim of re-commissioning the beam line to improve its performances (+50% beam transmission) with both simulations and experimental activities. - HIE-ISOLDE Project. Extensive campaign of simulations to support the experimental measurements of the longitudinal properties of the beam after the REX RFQ. - Medical Accelerators. Consulting for TERA on the design of the patients beam delivery system for linac based cancer treatment centres. TEACHING: - Supervised 2 Master students working on the design of medical machines - Trained 2 exchange students

Consorzio RFX | Jan 2008 - Sep 2009

Research Assistant

RESEARCH: - CNAO Project. Commissioning of the linac injector at the Centro Nazionale Adronterapia Oncologica (CNAO) in Pavia (Italy). Complete characterisation of the low energy beam transport lines, the RFQ, the IH-DTL and the stripping foils station in day and night shifts. Hundreds of measurements including transmission, energy with Phase Probes and emittance reconstruction as function of the machine parameters. - EURISOL Project. Prepared the proposal for the EURISOL post-accelerator injector beam dynamics. This included two RFQ beam dynamics designs (Normal Conducting and Super Conducting), a matching line and a solution for the beam pulsing using a multi-harmonic system (code: Mathcad). TEACHING: - Lab demonstrator / assessor for the Department of Engineering of Padua University (Italy) Outreach: ► Member of the LOC of the international conf. on Heavy Ion Accelerator Technology (HIAT’09) ► Editor of the Proceedings of HIAT’09

LNL-INFN | Jan 2005 - Dec 2007

PhD Student

RESEARCH: - Leading the beam preparation group as a beam physicist to improve the performances of ALPI, the Independently Phased SC Cavities Linac of LNL (+30% beam on target). - Commissioning the PIAVE (S-C injector) while being involved in intense experimental activity (machine setup, emittance measurements, data analysis). - Beam studies for the use of stripper foils in ALPI for ion charge enhancing and outlined some feasible upgrades of ALPI for oncoming LNL projects. - Experimental characterisation of ALPI and its injectors, the results of the ECR Ion Source Acceptance Test, describing the advantages of using a stripping station along the linac and the studies for the update/upgrade of the linac lattice for the SPES Project. - Collaborated on the development of the DTL for the early design of the SPES project based on the Linac4 DTL design. In addition, I participated in the studies for the beam transport lines, fitting the new accelerator structures into the present LNL buildings.

LNL-INFN | Dec 2003 - Nov 2004

1 year MSci Fellowship

1 year fellowship to complete the work I started for my Thesis. - Design of the Medium Energy Transport Line of the SPES Project (code: Parmila). - Optimisation of the shape of the high power Beryllium target for the BNCT facility. - Estensive error studies campaign to test the robustness of the solution (code: Fortran).

DESY | Jul 2002 - Aug 2002

Summer Research Student

Summer research placement at HASYLAB (Hamburger Synchrotronstrahlungslabor, "Hamburg Synchrotron radiation Laboratory"), working on the analysis of crystals. Programmed (C) a thermal probe to provide a real-time visualisation of the temperature next to the experiment and create a log of temperatures to be associated with different type of measurements.

My Education

Take a look at my formal education

Doctor of Philosophy (Ph.D.), Accelerator PhysicsUniversità degli Studi di Padova | 2008
Master, PhysicsUniversità degli Studi di Padova | 2004

Powered by

  • Work
  • Courses
  • Experience
  • Education
  • Create Your Data Portfolio for Free