Resume
Academic Background
-
Master Degree
- Domain: AI
- Politehnica University of Bucharest
- Faculty of Automatic Control and Computers
- Focus: Unsupervised Domain Adaptation
- Courses: Deep Neural Networks, Adavanced Techniques in AI, Computer Vision, Knowledge Representation and Reasoning, Multi Agents Systems and Mathematics for CS.
-
Bechelor’s Degree
- Domain: Computer Engineering
- Politehnica University of Bucharest
- Faculty of Automatic Control and Computers
- Focus: AI / Generative Models
- Thesis: Generative Adversarial Networks for Time Series Forecasting.
Work experience
Orange Services
- Data Scientist - January 2023 - Present
-
Responsible for creating and maintaining automatic speech recognition systems in GCP and NLP heavy load solutions (Named Entity Recognition, Dialogue State Tracking, Automatic Speech Recognition)
- Create ASR benchmarks for different technologies.
- Working on improving the performance of speech-to-text models for Romanian Language with model and dataset quality improvements.
- Create the production environment for heavy load models like OpenAI Whisper, custom models etc.
- Resolve the challanges in MLOps ( ASR, NLP) to keep the inference cost low: model quantization, batch inference etc.
👀 Mentoring for Bechelor Thesis projects and Disertation projects as company representative
- Network topology generation using Genetic Algorithms (Bechelor Thesis)
- Predictive analyze of the network data (Disertation)
- Techniques for network data forecasting (Disertation)
🎙 I am also actively participating in Orange Services Bootcamp, teaching an Intro to AI class for 3rd year students at the Politechnica University of Bucharest and many other events as speaker (Job Fairs, Awerness campaigns about the company and AI in general)
-
- Data Engineer - May 2022 - January 2023
-
I have started a Research Reading Group in which I am responsible for the general direction for each session, present the state of the art and implement different models as alternatives for our business cases solutions. 📰
-
Responsible for end to end GCP data migration pipelines implementation
- Assist the Data Science Team on integrating the models
-
Working alongside the Research team from France on developing a Dialog State Tracking end to end model as part of my diseartion thesis.
- Working on improving the performance of speech to text models for Romanian Language with model and dataset quality improvements.
-
Facultatea de Automatică și Calculatoare, UPB
- Graduate Teaching Assistant - Sep 2021 - Aug 2022
- Part of the Artificial Intelligence Lab.
- Held AI labs & seminaries over the semester.
- Prepared homework assignments and graded them.
- Prepared exam subjects (practical and theoretical) and graded them.
- Part of the Artificial Intelligence Lab.
- Undergraduate Teaching Assistant - Sep 2020 - Sep 2021
- Part of the Formal Langauges and Automata Theory department.
- Held Formal Languages Theory seminaries over the semester.
- Prepared homework assignments (i.e.: a custom language interpreter for Glypho programming language).
- Graded exam submissions and actively participated in curriculum planning.
- Part of the Formal Langauges and Automata Theory department.
Stealth Startup (Undisclosed)
- Software Engineer (Full Stack) - Jan 2022 - May 2022
- Working on a full stack fintech application.
- Integrated crypto exchange APIs into the platform
- Created and maintained the front-end app
- Integrated different features into the customers flow in order to enhance the user experience
Wave Motion Labs
- Software Engineer - Mar 2021 - Jan 2022
- Developed a real time Big Data visualisation web application that dynamically renders heterogeneous data in under one second.
- Coordinate a small front-end engineering team and integrate new features in large scale React application.
- Developed a customer event ontology in order to track user behaviour on various web platforms based on Apache Kafka streams.
- Developed data cleaning and data engineering tools for Computer Vision pipelines based on Apache Airflow.
- Worked on SSR solution for ecommerce platforms based on Jamstack & Gatsby & GraphQL.
NXP Semiconductors
- Software Engineer Co-op: Nov 2019 - Nov 2020
- Create / integrate new drivers / existing drivers into the core of Zephyr OS
- Create testing environment for specific sample applications using Synopsys VDK, TRACE32 ( debugger ) and networking concepts
- Working alongside the OS Team for extending the functionalities offered by NXP for Linux Automotive distributions to Zephyr
- Main programming languages and tools: C, Python [ west ] and Configuration files
Hootsuite
- Frontend Developer Co-op: Jun 2019 - Sep 2019
- I took part in the process of development of a real tool that will help the front-end team from Bucharest to speed up the React coding.
- Most of the time I worked with React & TypeScript and JavaScript for automation scripting.
- Create Jenkins cron jobs using Docker
- The main project consisted of a React Component that uses Cosmos( server like front-end development tool which uses a large amount of proxies to separate different components in a React project to make them accessible for developers) that acts as an Icon store with copy and paste feature.
- Main programming languages and tools: TypeScript, React, Docker, Jenkins
Medic Chat
Medic Chat is a Romanian Start-up in e-Health that has the purpose to intermediate the process of classical consultations by using an online portal where medics can answer questions and can give an advice.
- Web developer Co-op
- Work with the founders at different strategies for improving the monthly number of active users and customers - Develop front-end for the main application using Angular 6
- Automate the email process for new customers
- Automate the testing process of the web application using Python and Selenium
- Main programming languages and tools: TypeScript, Angular, Python