My name is Silviu Daniel Eftimie, I’m a Madrid-based Principal Cloud Developer at Mediabrands, a life long learner by heart, and a proud dad (1|2 ).
I like books (paper format) - they are my daily dose of inspiration. Sometimes I write. I keep in touch and let myself be guided by a number of people.
Retrospectively analyzing, when I was 4 years old I used to express myself through building. This became a passion. Later I understood that passion cannot be nurtured without a principle. When I realized that as a programmer you don't need anything more than a computer to create, I began to convert ideas into digital tools that allow people to understand and solve problems.
- Short-term: finding a working environment able to assess, develop and use my skills, creativity and knowledge that I have accumulated over time. I think that spending a great deal of your life at work and choosing energy in where and who you work with is always a good investment.
- Long-term: developing a digital mentor.
Madrid Area, Spain (Dec 2020 - present)
- Design, Develop and Deploy highly scalable and reliable Cloud Native Applications on GCP;
- Automate boring stuffs (internal and clients processes);
- Collaborate with all agencies and teams from Mediabrands;
Madrid Area, Spain (Jan 2016 - Dec 2020)
- designing data pipelines (ETL / ELT) to ingest, store, transform and activate data (GCP: Dataflow, Bigquery, Google Storage, Cloud Sql);
- designing data processing solution in batch and real time (with Cloud Functions, Cloud Run, PubSub, Dataflow, BigQuery);
- migration/creation/maintaining of data lakes and data warehouses;
- assessing the current state of a data warehouse;
- designing the future state of a data warehouse;
- migrating data, jobs, and access controls;
- deploying ML Pipelines (with BigQuery ML, AutoML Tables, Platform AI);
- measuring, monitoring, and troubleshooting pipelines and processes (Cloud Monitoring and Cloud Logging);
- leveraging prebuilt machine learning models as a Service (AutoML, Google ML APIS);
- developing dashboards with Data Studio / Tableau (few, but I have done this job too);
- cleaning data (using Dataprep, Data Fusion, custom code or our internal platform);
- scheduling and managing workflows (Cloud Composer / Data Fusion / Crmint, Cloud Schedule);
- SaaS development platform: I’ve built a visual data workflow orchestration service, serverless, multi cloud platform (gcp, aws, azure and on premise) capable of ingest, transform and process data at scale without writing code (although for custom jobs, you can use plain python/bash/sql);
- SaaS development platform: making valuable content accessible to international audiences using ML;
- SaaS development platform: setup programmatic advertising ads and monitoring KPIs from different DSPs (sources);
- automated processes: Invoice parsing tool / Generate invoice platform / Alert clients for Late Payments;
Madrid Area, Spain (Jan 2015 - dec 2015)
- implemente, setup and test analytics SDKs in mobile apps for understanding: impact of ads, attribution, user behaviour for understanding better the audience, improving user experience and client apps on Android and iOS. I’ve used firebase, facebook parse, adobe, google analytics, appsflyer;
- talks with clients (to understand better they're problems) and provide as quickly as possible a good solution;
- collaborating with clients developers;
- automated processes to ingest, store, transform and activate data in real time and batch;
- integrations via API of lots different systems; (eg: appsflyer to client data warehouse)
- collaborate with many people, different projects for Havas Media Agencies: Arena, DBi, Ecselis, Forward, Socialyse;
- A/B Testing on web and mobile;
Madrid Area, Spain (Dec 2009 - Dec 2014)
- SaaS development platform: inhouse CRM (monitor performance of teams, big picture on every clients);
- SaaS development platform: P&L Management System
- integrations via API of lots different systems;
- implemente, setup and test our JavaScript solutions for serving ads at scale on web platforms;
- designing data pipelines (ETL);
- automate reports;
Bucharest, Romania (Jan 2007 - Mar 2010)
- talk with clients;
- code & design for many websites (eg: https://www.pentruacasa.com);
- deploy and maintain the solutions;
School: Romanian-American University (Bucharest, Romania) Title: Bachelor in Cybernetics, Statistics and Economic Informatics (2006-2009)
School: Technical College Constantin Istrati (Campina, Romania) Title: Analyst programmer, Mathematics and Computer Science (2002 - 2006)
- GCP Professional Cloud Developer (see credentials)
- GCP Professional Data Engineer (see credentials)
- DeepLearning AI Tensorflow Developer (see credentials) (see my struggle)
- Facebook Marketing Developer (see credentials)
- Advanced Google Analytics (see credentials)
- Google Tag Manager Fundamentals (see credentials)
- Tealium iQ Advanced Deployment & Solutions Certification
- GTM Server Side Certification (see credentials)
- Google Analytics Individual Qualification (see credentials)
- Introduction to Data Studio (see credentials)
- Winners (Best in Show) Google Cloud - Easy as pie serverless Hackathon 2022
- MLess at Scale on GCP Blog
- MLess at Scale on Google Dev Library
- Prometeo, Google Marketing Partners Awards '21
- Programmatic Talks, Google Cloud Madrid (europe-southwest1!)
- Debate Mate ("Easy as pie serverless")
- GMP feat GCP: Firebase Hosting, GTM, GA4 and BigQuery
- Kick Off de Mediabrands 2021 (post | talk)
- Superautopista (post | talk)
- Tecnologías de Big Data al Servicio del Marketing Digital" via Edix
- As a Google Cloud Professional Data Engineer, I've presented for different clients all the Google Cloud Solution for Big Data and Machine Learning;
- In august 2020, I founded TF Team / Group: 14 members, from 5 different companies with the main scope to learn together TensorFlow and pass the exam. For more details please consult: First breakthrough, my progress with TensorFlow in Practice Specialization
- Talk at BMind, “Firebase - A game changer” collaboration with Christian Stoyanov;
- Cloud: GCP, AWS, Azure;
- Programming languages: Python, PHP, Node JS, Javascript, Java, Ruby, Flutter
- Frameworks: laravel, wordpress, express, spring boot, rails, vue, lumen
- Platforms: Firebase (all stack), Heroku, Salesforce Marketing Cloud, GTM, Tealium,GA
- Big Data: Cloud Functions, PubSub, Dataflow, BigQuery, Google Storage etc
- AI: Platform AI, AutoMLs, Kubeflow
- Tag managers: GTM (client and server side), Tealium (client);
- Privacy solutions: based on iAB TCFV1 and iAB TCFV2 specs;
- Digital marketing automation: using Salesforce Marketing Cloud + Firebase + GCloud / Heroku;
- Some DSPs & AdServers solutions;
- Data warehouses (migrate several projects from premise to GCP);
- ETL pipelines & processes in cloud;
- Mobile Apps (using Flutter, Swift, Java)
- Web platforms (Vue, Node JS, Laravel, Tailwindcss)
- REST Api integrations
- Smart Analytics Solutions for Big Data in Cloud;
-
SAS (Smart Analytics Solutions): using several Google Cloud Platform products, I have managed to set up a serverless, measurement solution in cloud (using Cloud Functions/Cloud Run + PubSub + BigQuery and some solutions from open source community: Great Expectations, Snowplow, StreamProcessor).
-
YTA: A visual data workflow orchestration service, serverless, multi cloud platform (gcp, aws, azure and on premise) capable of ingest, transform and process data at scale without writing code (although for custom jobs, you can use plain python/bash/sql). By myself, I have created the solution working on all its aspects: frontend, backend, architecture, deployment. In Digital Marketing it’s useful for cases like: Easy integrate your cloud applications or multiple systems; Automate marketing, sales and business processes without writing code; Transformation and pre-processing of data at scale using cloud tech; Supports many integrations & connectors
-
ML Solution: I've designed and worked for RTVE, using Google Cloud ML solutions : "Making valuable content accessible to international audiences". Case study here.
-
Banking: Abandoned cart (pre-Granted Loans solution): in real time detect and inform the Call Center about any user when they exit the process of contracting a loan without having finished it. The Call Center connects in less than 1 min with the users, assisting them for completion of the process. My part was to integrate the Call Center with Salesforce Marketing Cloud (sending them all info, in real time). In the past, the client would generate a report every day with users who have not finished the process (the day before) and they would pass that report to the call centers.
-
Media: Digital Asset Management: I've built an AI-Powered Searchable Video Platform Archive using Google Cloud Machine Learning solutions (Natural Language, Video AI, Speech to Text, Cloud Translation). The platform would analyze in parallel lots of video, extracting & storing metadata and generating subtitles.
-
Teachable Machine 4 Youtube/Netflix (code): Open source demo, using machine learning to play/pause videos from distance on Youtube, Netflix or HBO in a hacky way. I've been using Teachable Machine (from Google AI Experiments), some Javascript and Chrome Developer Tools. Now, this project it’s part of “The Awesome Teachable Machine List”
-
Rock Paper Scissors with Google Teachable Machine (game | video | code): I have used again the magic of Google to train a visual machine learning model and turn everything into a classic game for my kids: Rock Paper Scissors.
- Making Friends with Machine Learning
- The Infinite Game
- The downside of sir
cancherity
- Book review: Building Serverless Applications with Google Cloud Run
- Decoding Decisions - Making sense of the messy middle
- Apache Beam / Dataflow
- Welcome GCP Madrid (europe-southwest1)
- Application Development with Cloud Run
- Unthinkable vs Inexpressible Thoughts
- How to become a data-driven retailer
- Google Marketing Livestream 2021
- Extend Google Analytics with Firebase Cloud Functions
- Adobe on cookieless
- BigQuery as a CDP?
- GMP + GCP fusion
- On mentorship (A Mind at Play: How Claude Shannon Invented the Information Age)
- Firebase Summit 2020: Import custom segments
- Add Recommendations to your app with Firebase ML, Google Analytics, BigQuery
- Peter Norvig on "As We May Program"
- GCP Resilient Systems / Buildpacks (Google Cloud Next)
- GCP: Stream Analytics
- Streamlining Cloud Run development with Cloud Code
- Learning and contributing to open source projects;
- Education: deeply involved in the education of my kid. (eg: Chess for kids, Coding for kids)
- Sharing ideas, solutions and connecting with lots of people (linkedin, twitter, github)
- Tech: DevOps, MLOps, FaaS, CaaS paradigms. Machine Learning with Tensorflow. Data multi cloud solutions; (eg: Anthos, BigQuery Omni). Solution for big data / validation and cleaning data. Mobile and web dev using Flutter;