Jewish Digital Cultural Recovery Project (JDCRP) Public Tender
OCR Expert for AI-Driven Text Recognition and Digitalization Support
1. Background
The Jewish Digital Cultural Recovery Project (JDCRP) is a non-profit foundation based in Germany, dedicated to documenting and reconstructing the history of Nazi-looted cultural property. The JDCRP is developing a centralized digital platform that aggregates archival data, enhances provenance research, and provides historical insights into the looting of Jewish-owned art and cultural property by the Nazis.
As part of this effort, JDCRP is seeking an OCR (Optical Character Recognition) expert with experience in AI-driven text recognition solutions. This expert will play a key role in developing and improving OCR capabilities, ensuring that archival materials in various formats, languages, and conditions can be successfully digitized and integrated into the platform.
This tender is part of two parallel projects:
A project funded by the German Federal Government Commissioner for Culture and the Media (BKM)
A second project supported by the European Union (EU) and the Conference on Jewish Material Claims Against Germany (Claims Conference)
The selected expert will start immediately, with the possibility of continued collaboration.
2. Scope of Work
JDCRP is looking for an expert to:
Develop and implement OCR solutions, ideally using Generative AI (Gen-AI) technologies, to process complex archival records in multiple formats and languages.
Evaluate and test OCR approaches for different types of archival documents, considering variations in quality, structure, and historical context.
Support JDCRP researchers and data officers in reviewing records and selecting the best OCR strategies for digitization.
Refine and improve OCR accuracy for historical documents, handwritten materials, multilingual records, and low-quality scans.
Participate in internal workshops and discussions to help define future OCR strategies and contribute to the next phases of JDCRP’s digitalization efforts.
Ensure seamless integration of OCR solutions into the JDCRP platform and provide recommendations for scalable, long-term approaches.
3. Required Qualifications
The ideal candidate should have:
Proven expertise in OCR technologies, with experience in AI-driven OCR and Gen-AI solutions.
Experience handling complex archival records, including historical documents with different structures, languages, and conditions.
Strong problem-solving skills and ability to develop custom OCR solutions for diverse document types.
Ability to collaborate with both technical and non-technical teams, including provenance researchers, historians, and data officers.
Excellent communication skills and flexibility to adapt to emerging project needs.
Experience working with cultural heritage, historical archives, or research platforms is a plus.
4. Project Timeline & Engagement
Start Date: Immediate
Initial Engagement: Short-term with an opportunity for continued collaboration
Location: Remote or hybrid (with occasional in-person meetings if required)
5. Bidding Process & Submission Guidelines
Interested candidates should submit:
A short proposal (1-2 pages) outlining their experience and approach to OCR development for archival materials.
Examples of past OCR projects, particularly those involving historical or multilingual documents.
Proposed hourly or project-based rate.
Availability and estimated time commitment.
Deadline: Rolling applications accepted (as soon as possible).
Send applications to: info@jdcrp.org
6. Additional Information
Questions & Clarifications: Candidates may request additional details by contacting JDCRP at info@jdcrp.org
Evaluation Criteria: Proposals will be assessed based on technical expertise, OCR experience, cost-effectiveness, and ability to collaborate with JDCRP teams.
Contract Terms: JDCRP reserves the right to negotiate contract terms with the selected expert.
We encourage qualified professionals to apply as soon as possible.
JDCRP Public Tender WP2: Data Linking and Matching
The JDCRP, a non-profit foundation based in Germany, is creating a digital platform for archival documentation related to the art looted by the Nazis, their collaborators and allies. The platform will allow for cross-searching of documents from different archival sources. The European Union co-funds the two-year project “JDCRP Pilot Project Part II” together with the Conference on Jewish Material Claims Against Germany (the Claims Conference), sponsored by the Foundation “Remembrance, Responsibility and Future” (EVZ) and supported by the German Federal Ministry of Finance (BMF). A second project for the development of a JDCRP Pilot-Minimum Viable Product with co-funding by the German Federal Government Commissioner for Culture and the Media (BKM) and the Claims Conference.
To achieve these goals, the JDCRP is inviting bids for a web application that supports the creation of Linked Open Data by ingesting data sets and connecting them among each other. In addition to interconnecting datasets, other open data sources such as thesauri and vocabularies must also be used to further enrich the data. The application will make the linked data available to a front-end system that will be tendered separately. The archival material identified in this project for inclusion on the platform is highly diverse in its format, content, structure, and state of digitization.
JDCRP Pilot Project Part II
As a first step, JDCRP aims to make available and searchable on its platform a set of 6,673 property cards that were generated at the Wiesbaden Central Collecting Point; these cards have been scanned but are not yet searchable. Once the contents of the cards are available digitally, the goal of the requested service is to ingest them, together with existing data from databases on the subject such as the ERR Jeu de Paume database, and create connections between the two sets by matching them through a variety of strategies. Other examples of search challenges might be file identifiers that are matched with items by their identifiers. In addition to these original data sources, smaller sections within the platform will be created to make available in-depth research on two art collections and on an educational project that are developed within this project.
JDCRP Pilot Minimum Viable Product
The goal of the second project is to prepare, ingest, and make archival material from multiple sources cross-searchable on a central platform. This is done to create and elaborate the structure, interconnections, and searchability of the platform. These sources may include property cards from other Central Collecting Points, as well as adding new layers of data to the platform, such as scans of photographs that are available at the U.S. National Archives and Records Administration (NARA). In addition, research indexes on, for example, persecuted Jewish artists and collectors, must be ingested as datasets. At this stage, it is furthermore anticipated that the material and data that will be available in the central data platform will be interconnected with the other sections of the overall platform that contain research and educational material.
Financing
This tender includes two sets of activities that are part of two different projects. As these projects are financed by different sources, the JDCRP will require two different bills for the two sets of activities. Apart from this tender, the JDCRP will also publish a tender for the front-end in the near future. At the end of the project, we also request a summary report explaining the challenges and solutions found. The successful bidder is required to begin development as soon as possible. The end date for the development work is set for October 30, 2024, however, the service should be available beyond this time frame. We furthermore expect a successful service provider to adhere to modern software standards and that time for testing and iterative improvement is allocated throughout the duration of the development.
Bidding process
In your bid, please include the following information:
- Describe how each of the work packages will be approached and which technology will be used.
- Provide a cost estimate for each service offered that includes a detailed specification as well as the number of persons and/or work hours involved.
- Outline a time frame and schedule you envision being needed in order to set up the service.
- Explain how you ensure quality management according to the team´s aims.
- Describe your company’s track record dealing with similar projects, preferably in cultural heritage, provenance research, and/or art history.
- Include the individual qualifications of the people who will participate on the project.
- Estimate the cost of the maintenance and/or other costs expected for the continued availability of the service.
The tender project information is available here, and the full tender package can be requested here: secretary@jdcrp.org.
Bids should be submitted by December 15, 2023, to:
Jonathan Blok, IT / Data Officer: jonathan.blok@jdcrp.org