Description

You will design and implement AI-Native agentic systems to extract insights from historical and genealogical records.

Responsibilities

• Implement state-of-the-art AI solutions for Document Understanding tasks, including OCR/HTR, NER, and Relation Extraction.
• Architect multi-agent workflows using frameworks like LangChain and AutoGen to automate complex reasoning in document analysis.
• Evaluate multi-modal models (Gemini, Claude, GPT, Qwen) for zero-shot and few-shot document understanding performance.
• Establish ensemble models and LLM-as-a-Judge frameworks to monitor for hallucination and bias.
• Partner with ML Ops to deploy models and pipelines in cloud environments like AWS and GCP.

Required Skills

• 10+ years of experience in a quantitative field such as Computer Science or Data Science.
• Specialization in AI and LLMs, including familiarity with foundational models like GPT, Gemini, and Qwen.
• Strong proficiency in Python and relevant libraries, including transformer and multi-modal models.
• Experience designing and implementing agentic systems using frameworks like LangChain, LangGraph, CrewAI, or AutoGen.
• Expertise in NLP tasks such as NER, Relation Extraction, and Knowledge Graphs (Neo4j).
• Familiarity with computer vision models like YOLO, Nougat, or OpenCV for layout analysis.
• Experience with inference optimization techniques such as LoRA and quantization.
• Experience developing and deploying services on cloud platforms, specifically AWS and GCP.
• Ability to present complex technical solutions to both technical and non-technical audiences.

Key Skills

Education

Any Graduate