About the Role
Title: Senior NLP Engineer (AI Product) (Remote)
Location: worldwide
Category: Software Development
Job Description:
Social Discovery Group (SDG) is a global technology company that builds apps at the intersection of dating, social, and entertainment. The company’s portfolio includes 70 social discovery platforms with a focus on AI, game mechanics, and video streaming. We actively support and invest in social discovery startups worldwide through our CVC fund.
More than 500 million people in 150 countries enjoy our products, and we strive to have ten times as many.
SDG invests in social discovery technology startups around the world. Our Investments include Open AI, Patreon, Flo, RAW, EVA AI, Clubhouse, Magnet, Woebot, Flure, Astry, Coursera, Academia, Harbour, Space, Auto1, DocSend, AppAnnie, Rapyd, Boom Supersonic, Trading, View, K-Health and many others.
We solve the problem of loneliness, isolation, and disconnection with the help of digital reality.
Our digital nomad team of more than 1200 professionals works all over the world. Together, we are solving the prevalent problem of loneliness and shaping Social Life 3.0 – a new digital reality where people will be able to fulfil their needs for communication and attention from other people and artificial life forms.
Our teams of digital nomads live and work remotely from Cyprus, Malta, the USA, Thailand, Indonesia, Hong Kong, Japan, Australia, Poland, Israel, Turkey, Latvia and many others.
We are looking for a Senior NLP Engineer to join our new product, TheOne.AI.
TheOne.AI is an artificial intelligence platform created by the Social Discovery Group. It is designed for natural communication and entertainment and helps avoid feelings of loneliness.
TheOne is an AI-driven platform designed for authentic interaction and entertainment, helping to combat loneliness. It features a unique profile marketplace with detailed backstories, covering individuals of all ages, genders, and nationalities, each with their unique interests. AI bots behind these profiles offer a personalized communication experience, mimicking real human interactions so convincingly that the line between artificial and real fades away.
These bots can engage in conversations, send voice messages, and share photos and videos, creating a sense of genuine interaction. TheOne allows users to forget they’re interacting with AI, offering a deep dive into the world of emotions and interpersonal relationships. It redefines virtual entertainment standards, making every interaction unique and rich, setting a new benchmark for AI developments.
Your main tasks will be:
- Conducting experiments with LLMs: Explore and analyze the effectiveness of different architectures and techniques (SFT, RLHF, Adapters, etc.) to enhance the capabilities of AI models;
- Developing and implementing evaluation methodologies: Implement and maintain robust frameworks to assess the performance, accuracy, and user satisfaction of AI bots, including offline and online metrics;
- Optimizing models for inference: Improve the efficiency and speed of AI models during inference to ensure they meet the performance and scalability requirements for production environments;
- Collaborating with cross-functional teams: Work closely with data scientists, software engineers, and product managers to integrate AI solutions into the overall product pipeline.
We expect from you:
- Proficiency in Python, algorithms, and mathematics: Strong coding skills in Python and a solid foundation in algorithms, linear algebra, probability, and statistics;
- Deep understanding of ML and DL principles: Comprehensive knowledge of machine learning and deep learning concepts, including supervised, unsupervised, and reinforcement learning;
- Comprehensive expertise in NLP and LLM: In-depth experience with natural language processing and large language models;
- Familiarity with ML frameworks and tools: numpy, pandas, scipy, scikit-learn, pytorch, transformers;
- English level: B2+.
It will be a plus:
- Experience in distributed training: Ability to implement and manage distributed learning workflows, ensuring efficient parallel training of models across multiple GPUs and nodes;
- Proficiency in C++ and CUDA: Strong skills in C++ programming and experience with CUDA for optimizing and accelerating deep learning algorithms on NVIDIA GPUs;
- Relevant publications: Demonstrated expertise through publications in reputable conferences or journals, showcasing contributions to the field of machine learning and AI;
- Knowledge of cloud platforms: Hands-on experience with cloud services like AWS, GCP, or Azure for training DL models.