Artificial Intelligence Engineer

droppGroup is hiring an exceptionally capable AI Engineer with deep, practical expertise in generative AI and natural language processing (NLP). This role is not for generalists, we are looking for specialists who can operate across the full AI lifecycle: from research implementation to real-time production deployment. You must bring a disciplined engineering mindset, state-of-the-art awareness, and hands-on experience with training, fine-tuning, and deploying large language models at scale. Responsibilities Design and fine-tune large transformer models for complex generative tasks including chat, summarization, and semantic understanding. Build LLM-based systems using advanced prompt engineering, retrieval-augmented generation (RAG), and optimized inference. Own the development and optimization of model training pipelines using PyTorch, Hugging Face, DeepSpeed, and related frameworks. Deploy and maintain LLMs in high-performance production environments, optimizing for latency, cost, and stability. Conduct rigorous evaluation of models using benchmarks, human feedback, and automated testing. Manage end-to-end AI workflows with robust versioning, reproducibility, and observability using MLOps tooling. Implement scalable vector databases and semantic search infrastructure for embedding-based retrieval systems. Translate latest AI research into efficient, production-grade code—no research-only profiles. Size, allocate, and optimize compute resources (CPU/GPU/TPU) for large-scale training and inference. Establish strict testing, rollback, and fail-safe mechanisms for model deployment in live systems. Qualifications Minimum 3 years of focused experience building deep learning systems, with a proven track record in NLP and/or generative AI. Hands-on experience training and deploying transformer-based models (BERT, T5, GPT, LLaMA, etc.)—from scratch or via fine-tuning. Expert-level Python, with production-grade PyTorch skills (Tensor manipulation, gradient debugging, memory profiling, etc.). Demonstrated ability to work with multi-billion parameter models, and deploy them under real-world latency and throughput constraints. Deep understanding of language modeling, embeddings, attention mechanisms, tokenization strategies, and architectural tradeoffs. Ability to read, implement, and extend research papers into robust, testable, and efficient systems. Experience with AI infrastructure at scale, including distributed training, mixed precision, checkpointing, and A100-level GPU scheduling. Familiarity with vector databases (FAISS, Weaviate, Milvus) and hybrid search systems. Proficiency in MLOps tools (MLflow, Airflow, WandB, or similar) for model lifecycle management. Strong commitment to software engineering principles: code quality, version control, CI/CD, and modular design. Capacity to own AI systems end-to-end—from dataset preprocessing to online deployment and continuous improvement. droppGroup is an equal opportunity employer. We offer a competitive compensation structure and a fast-paced growth roadmap to all our teams. Show more Show less