Looking for a job

Benjamin Ciaglo

@benjaminciaglo

Lead Software Engineer at ITHAKA (JSTOR) who developed JSTOR Seeklight, an AI document platform processing 210,000+ documents.

United States

Message

At ITHAKA (JSTOR), I developed JSTOR Seeklight, an AI-powered document intelligence platform within the JSTOR Stewardship Digital Asset Management system. It has processed 210,000+ documents and produced 90,000+ transcripts for 60+ institutions, and won the Society of American Archivists’ 2025 C.F.W. Coker Award for Description.

I architected its document AI and RAG pipelines, built production services, and led cross-functional engineering work to integrate Seeklight with Forum. Earlier, I developed AI and software systems at Appen, Amazon, Portal 724, Ford Motor Company, and other organizations; I also hold an M.S. in Computational Linguistics.

Experience

Work history, roles, and key accomplishments

Lead Software Engineer

ITHAKA

Jan 2025 - Nov 2026 (1 year 10 months)

Technical lead for JSTOR Seeklight, an AI-powered document intelligence platform, leading from foundation through beta and first-year integration. Architected multi-step AI pipelines and led cross-functional teams of ~6 engineers.

Python NodeJS Vue.Js AWS Lambda Step Functions API Gateway Terraform PostgreSQL OpenSearch RAG LLM Evaluation Machine Learning Natural Language Processing

Senior Software Engineer

ITHAKA

Nov 2022 - Jan 2025 (2 years 2 months)

Founded and delivered JSTOR Seeklight, leading end-to-end beta development. Backend engineer on Forum and full-stack contributor to JSTOR's AI/LLM-enabled research assistant.

React Python PostgreSQL DynamoDB OpenSearch RAG Kubernetes Clojure Ruby LLM Evaluation

Senior Computational Linguist

Appen

Jan 2022 - Sep 2022 (8 months)

Developed and delivered React and Python features for a Gremlin/Neptune linguistic tooling POC, transitioning it from development to customer-ready status within six months.

React Python Gremlin Neptune AI Training

Software Development Engineer II

Amazon

Jun 2020 - Mar 2021 (9 months)

Upgraded Spark-based data processing platform running 1M+ jobs/month and designed Java APIs for permissions preflight check used by thousands of internal users.

Spark Python Java CI CD API Design

Full-Stack Developer

Portal 724

May 2017 - Mar 2020 (2 years 10 months)

Built Python backend systems detecting abnormal device usage, reducing data costs by $70,000+ per month. Refactored production applications using Ionic, C#, and SQL, improving load times by 90%+.

Python Ionic C#SQL