ABOUT THALES DIGITAL FACTORY
In an inauguration ceremony held November 29, Singapore became the third country, after France and Canada, to welcome the Thales Digital Factory.
Known for its technology investments, vibrant ecosystem and connectivity, Singapore is a prime destination for innovation to thrive. The addition of Singapore to the Digital Factory network will accelerate innovation and digital transformation in the region and allow Thales to leverage its innovative services in aerospace, space, ground transportation, defence and security to its customers in a collaborative way.
The Digital Factory was launched in July 2017 in Paris as a fully-fledged internal digital company whose mission is to accelerate the digital transformation of Thales and its customers. It was expanded to Montreal in April 2018 and already hosts more than 230 experts in digital technologies (Cloud, Internet of Things, big data, artificial intelligence, cybersecurity etc) and methods (design thinking, lean start-up, agile software development) to develop innovative products in close collaboration with future users. The Digital Factory delivers first versions of new digital services in a few months to quickly test operational and business value, through Minimum Viable Products (MVP). With 15 MVPs already delivered within one year in domains such as air traffic management, avionics equipment repair, drone operations, airport passenger experience, maritime traffic, metros maintenance, the Digital Factory allows Thales’s customers to quickly experiment new solutions at a rapid rate to improve their operations and customers experience.
“Hungry, Humble, Aware” – Patrick Lencioni
ROLE DESCRIPTION SUMMARY
As a Data Scientist in the Digital Factory, you would be expected to partake in solving machine/deep-learning problems in various domains (e.g. Avionic Systems, Radar Systems, Underwater Systems) with varying degrees of complexity. You should also have experience in presenting the results of experiments conducted, failures and successes towards key stakeholders in an understandable fashion. You should be someone whom has an advanced degree in Mathematics and/or Statistics. You should also have programming experience and understand how to create business savvy visualizations for your key stakeholders. You should also be someone whom has past experience coaching lesser experienced data science professionals. You should be someone whom has past experience in fine-tuning machine learning algorithms in production environments. You should be someone whom is aware that there is a lot you do not know but willing to acquire and validate knowledge; is not emotionally attached to share your failures and subsequent learnings. You should be someone whom is willing to partake in the day to day engineering activities of your team and factory.
KEY ACTIVITIES AND RESPONSIBILITIES
As a Data Scientist, you are accountable for:
• You will apply machine learning and possibly deep learning techniques to a variety of modelling and relevance problems involving our users, Thales Hardware, Thales Software with the end goal of delivering the AI solutions to production.
• You will participate in the engineering life-cycle at Thales Digital Factory, including daily scrum, sprint reviews, sprint retrospectives, community of practices etc
• You will have to review, regularly, for performance improvements and decide which AI technologies and algorithms can be used in a production environment
• You will partake in data exploration activities for businesses, uncovering patterns in the data usage.
• Extracting actionable insights from diverse data sources through data mining techniques
• You will design and apply algorithms to identify key features, build, and fine-tune models
• You will work closely with data engineer in implementing preprocessing of both structured and unstructured data
• You will evaluate and identify relevant datasets and create data dictionaries when applicable
• You will be handling data processing, cleansing, and validation to ensure its suitability for analysis
• You will deliver clear and concise presentations of analytical findings with stakeholders
• You will bring the best-in-class practices to the MVP team to make sure the data science is maintainable, scalable and debuggable
KEY KNOWLEDGE AND EXPERIENCE
To be successful in your role, you will have demonstrated and/or acquired the following knowledge and experience:
• At least 5 years of data science experience where you would have understood the process of negotiating and unraveling the nitty gritty details of your customer or user’s datasets with the end goal of building a demonstrable proof-of-concept i.e. PoC.
• You should have a good repertoire of software tools and programming languages to which you can apply to building a PoC.
• You should have good working knowledge about Machine Learning and/or Deep Learning algorithms in the realm of supervised, unsupervised, reinforcement not limited to ANN, CNN, RNN, GAN.
• You should have good working knowledge in statistical classification domain eg. Logistic Regression, K-nn, Kernel SVM, Naïve Bayes, Decision Tree, Random Forest.
• You should have good working knowledge in clustering domain e.g. K-means clustering, hierarchical clustering
• You should have a good understanding of applying gradient boosting in regression and classification problems. E.g. XGBoost
• You should have good working knowledge in anomaly detection and outlier detection techniques e.g. DBSCAN, Gaussian Mixture Models.
• You should have deep understanding and experience with large language modeling techniques, such as GPT, BERT, or Transformer-based architectures.
• You should be familiar with large language modeling frameworks like OpenAI's GPT, Hugging Face's Transformers, or Google's BERT.
• You should have knowledge of natural language processing (NLP) concepts and techniques, such as tokenization, word embeddings, and sequence modeling.
• You should have proficiency in reinforcement learning algorithms and concepts. This includes understanding the basics of Markov Decision Processes, Q-learning, policy gradients, and value iteration.
• You would have working knowledge of one or more programming languages like C & Python and you should be able to explain the underlying mechanics in addition to the theoretical Machine/Deep Learning model.
• You would have good working knowledge of the Machine Learning and Deep Learning toolkits available in OSS, commercial software.
• Worked in a squad or guild team setup and understand the agile processes, ceremonies and appreciates them
• Has a continuous learning mindset and learning of new programming paradigms, techniques & practices
• Open, strong communicator who communicates effectively across teams, locations and cultures, in-person and virtually
• Courage of convictions with a high degree of humility. Embraces constructive feedback and is resilient
