AI/ML Engineer

Biography

Highly motivated researcher and engineer with expertise in machine learning, program analysis, and information security. I completed a PhD in Dual-channel Program Analysis, leveraging the relationship between code and natural language to enhance code completion, vet type conversions, and assess test suite efficacy. Currently, I am working as an AI/ML Engineer at Hurdle, where I develop innovative AI solutions to address health challenges, including the early detection of diseases.

Interests
  • Machine Learning
  • Dual-channel Research
  • Program Analysis
  • Information Security
Education

  • PhD in Computer Science, 2024

    University of Surrey

  • MSc in Information Security, 2019

    University of Royal Holloway

  • BSc in Computer Science, 2018

    University College London

News

Presenting the Checking Test Suite Efficacy Through Dual-Channel Techniques as a General Track Research Paper at ICTSS ! See you there!
Talk Nov 2024
Started working at Hurdle as an AI/ML Engineer!
New Position May 2024
Presenting the Do names echo semantics? A large-scale study of identifiers used in C++’s named casts as a Journal-First Paper at ESEM ! See you there!
Talk Oct 2023

Recent Publications

Checking Test Suite Efficacy Through Dual-Channel Techniques
Constantin Cezar Petrescu, Sam Smith, Alexis Butler, Santanu Kumar Dash
PDF Cite Code
Do names echo semantics? A large-scale study of identifiers used in C++’s named casts
Constantin Cezar Petrescu, Sam Smith, Rafail Giavrimis, Santanu Kumar Dash
PDF Cite Code
Genetic optimisation of C++ applications
Rafail Giavrimis, Alexis Butler, Constantin Cezar Petrescu, Michail Basios,, Santanu Kumar Dash
PDF Cite

Experience

 
 
 
 
 
AI/ML Engineer
Hurdle
May 2024 – Present
Hurdle logo
  • Developing an AI-powered Health Insights Engine for healthcare companies to autonomously review personalised health risk scores for patients and prevent diseases.
 
 
 
 
 
Freelancer
Oct 2023 – May 2024
  • Trained and deployed LLM models (Llama, Mixtral) for diverse chatbot applications.
  • Implemented a multi-model LLM architecture, optimising app performance for specific queries.
  • Developed and deployed transformer models to detect hateful comments for chat moderation.
  • Developed sentence transformers and LLM models for semantic similarity.
 
 
 
 
 
Supervisor and Teacher Assistant
University of Surrey
Oct 2019 – Jun 2023
Surrey logo
  • Supervisor of Final Year Projects for BSc and MSc, and Research Summer Internships.
  • Teacher Assistant for COM2040 (Further Programming Paradigms) and COM1029 (Software Engineering).
 
 
 
 
 
Research Intern
UCL
Jun 2018 – Mar 2019 London, UK
UCL logo
  • Developed a neural network model to predict multiple tokens to improve code completion for C/C++.
  • Data gathering was done through NLP by building a Clang plugin to process open-source projects.
 
 
 
 
 
Software Engineer Intern
Emotech LTD
Jun 2017 – Oct 2017 London, UK
Emotech logo
  • Improved the architecture of existing scraper to allow multiple scrapers to gather data at the same time. Increased the speed of scraping by 30x and added functionality to make it easy to run and manage.
  • Developed 4 full-stack websites, migrated and improved functionality, UI and security.
  • Set up a set of servers on AWS capable of auto scaling according to the usage demand.
 
 
 
 
 
Software Engineer
Freelance
Jan 2017 – Jun 2017 London, UK
Developed a responsive website using agile principles. Front-end: HTML, CSS, JS, jQuery; Back-end: Node.JS, Express.JS; Database: MySQL; Data gathering was done using a Java web crawler.
 
 
 
 
 
Software Engineer Intern
Cazana
Jul 2016 – Oct 2016 London, UK
Gathered data from websites to compare prices of second-hand cars by developing set of scrapers and web crawler. Refactored the previous code, integrated the new one and deployed new versions.
 
 
 
 
 
Software Engineer Intern
Route1 LTD
Jun 2016 – Jun 2016 London, UK
Automated the data gathering of law job listings by developing scraping algorithm using Node.JS.