DeepLight AI is a specialist consultancy implementing intelligent enterprise systems, with deep expertise in financial services and banking. We are seeking a Generative AI Engineer to join our team, focusing on turning advanced research into high-impact, production-ready enterprise solutions.
You won't just train models; you will design, build, and own the entire lifecycle of AI systems that leverage Large Language Models (LLMs), RAG, and Multimodal AI to solve our clients' most complex problems.
The Challenge: Owning the LLM Production Pipeline
This role demands a unique combination of applied machine learning, data engineering, and scalable software delivery, operating entirely within a robust MLOps/LLMOps framework .
🛠️ What You Will Be Doing (Key Responsibilities):
- Design & Deliver GenAI Solutions: Leading the implementation of LLM-based applications, including custom chatbots, advanced summarization services, and creative co-pilots.
- Model Optimization & Steering: Applying cutting-edge techniques like fine-tuning, LoRA/PEFT, and RLHF to optimize model performance and ensure factual accuracy.
- RAG System Architecture: Architecting and building high-performance Retrieval-Augmented Generation (RAG) pipelines, managing the full lifecycle of embeddings and context-aware retrieval.
- Production Code: Writing robust, deployment-ready software in Python and TypeScript, delivering clean, modular code.
- MLOps Automation: Implementing full LLMOps/MLOps using Docker, Kubernetes, and Terraform to automate CI/CD, deployment, and versioning across major cloud platforms (AWS/GCP/Azure).
- Data Grounding: Ensuring model intelligence is fresh and reliable by connecting to Airflow, dbt, and Kafka pipelines.
- Governance & Safety: Embedding responsible AI practices, including hallucination control, bias mitigation, and auditability, into every deployed system.
