Job Description

• Develop and fine tune NLP/NLU models, including transformer-based models (BERT, GPT etc.,) for a range of language understanding tasks (language generation, sentiment analysis, entity recognition etc.,)
• Additionally design and implement generative AI models by leveraging LLMs for text generation and understanding.
• Preprocess and curate text data, including text cleaning, tokenization, and data augmentation to prepare it for training NLP/NLU models
• Innovate and implement cutting edge algorithms to improve the performance, accuracy, and efficiency of the NLP/NLU models
• Assess model performance using relevant NLP metrics and optimize models for real world applications
• Stay updated with latest developments in Generative AI specifically on NLP/NLU research and apply new techniques to enhance our solutions
• Strong expertise in NLP/NLU with focus on deep learning models and open source LLMs such as Llama 2, Falcon etc.,
• Proficiency in programming languages such as Python and experience with NLP libraries/frameworks like spaCy, NLTK, Hugging Face Transformers and Generative AI and LLM frameworks
• Familiarity with cloud computing platforms for model training and deployment
• Experience with tools like Langchain, Embedchain and Vector Database (Pinecone, FAISS, Milvus, ChromaDB, Vespa etc.,) to enhance data processing and model development workflow as well as rich understanding of vector stores and search algorithms
• Exceptional problem-solving skills and the ability to work on complex language understanding and Generative AI projects
• Experience designing control and sandboxing systems for AI research
• Hands on experience with Pytorch, TensorFlow
• Direct engineering experience of high performance, large-scale ML systems
• Hands on MLOps experience, with an appreciation of the end-to-end CI/CD process
• Have experience supporting fast-paced startup engineering teams
• Experience maintaining and/or contributing to bug bounty and responsible disclosure programs
• A well-known contributor to opensource and always thinking out of the box tooling, using and standardizing with methods of creating APIs , ML/Ops automation and more.