Postdoctoral Fellow: Applied AI for Document Understanding

Ancestry; United States; Hybrid; USD 110,000 - 160,000; Full time

Benefits offered by this job

Flexible work locations

Extremely collaborative environment

Job summary

Ancestry is seeking a Postdoctoral Research Fellow to advance AI for document understanding and autonomous multi-agent workflows. You will design, implement, and evaluate AI-native agentic systems to extract knowledge from billions of genealogical records and transform unstructured data into searchable insights.

You will lead research at the intersection of industry-scale data and academic rigor, collaborating across NLP, CV, and graph-based techniques, with a focus on scalable, real-world

Qualifications

• PhD in Computer Science, Data Science, Statistics, Linguistics, Engineering, or related quantitative field with strong research focus.
• Strong record of academic publications in NLP, CV, or Agentic AI preferred.
• Specialization in AI & LLMs, familiarity with models such as GPT, Gemini, Qwen, Llama, Claude.
• Research in inference efficiency and optimization; experience with vLLM, LoRA, QLoRA.

Responsibilities

• Innovate with state-of-the-art AI for Document Understanding tasks (OCR/HTR, NER, RE, Coreference, Summarization, Knowledge Graphs).
• Architect multi-agent workflows using LangChain, LangGraph, CrewAI, AutoGen, and related frameworks.
• Analyze and optimize multi-modal models like GPT, Gemini, Claude, Llama, Qwen for zero-/few-shot scenarios.
• Apply NLP (NER, relation extraction, entity resolution) and CV (layout analysis, text blocks) techniques.
• Collaborate with ML Ops to deploy datasets, models, and pipelines on AWS and GCP.

Skills

Python programming

NLP

Multi-agent systems

Transformer models

Knowledge graphs

Education

PhD in Computer Science or related field

Tools

LangChain

LangGraph

CrewAI

AutoGen

AgentCore

Strands

Neo4j

Job description

Ancestry is seeking a Postdoctoral Research Fellow to advance AI for document understanding and autonomous multi-agent workflows. You will design, implement, and evaluate AI-native agentic systems to extract knowledge from billions of genealogical records and transform unstructured data into searchable insights.

You will lead research at the intersection of industry-scale data and academic rigor, collaborating across NLP, CV, and graph-based techniques, with a focus on scalable, real-world