AI Engineer

We are looking for an AI Engineer to build and optimise advanced Retrieval-Augmented Generation (RAG) architectures and multimodal document processing systems. You will focus on extracting intelligent insights from complex unstructured data while managing costs and system scalability.
Responsibilities:
Design and implement RAG pipelines, handling large document chunking strategies, embedding generation, and vector database storage.
Develop layout-aware parsing and OCR solutions for complex documents, including multi-row tables and screenshots with highlighted regions.
Build temporal retrieval systems that effectively manage document version conflicts and historical metadata.
Orchestrate multi-agent systems, managing stateful agents, context windows, and mitigating token explosion.
Design cost-conscious AI architectures by integrating traditional NLP and rule-based pipelines when LLMs are not strictly required.
Requirements:
Deep expertise in vector search, similarity calculation (e. g., cosine similarity), and dense retrieval re-ranking.
Strong understanding of RAG evaluation metrics, including Context Precision, Context Recall, Faithfulness, and Answer Relevancy.
Experience with Context Generation and knowledge injection over basic prompt engineering.
Solid foundation in Classical Machine Learning, specifically tree-based models like Random Forest, Boosting, and XGBoost.
Strong foundational knowledge in probability, combinatorics, and quantitative problem-solving.

Experience
4-8 yrs