Web Scraping Developer Needed – iDoc Market Public Records & Document Data Extraction

Looking for someone to build a scraper/automation for Williams County, North Dakota records available through iDoc Market. The project will involve searching and scraping a large number of recorded documents across multiple instrument types. Some individual instrument types may return 10,000+ documents. The project has three main parts: Scrape the specified iDoc records Organize the records so they can be searched and reviewed by person Conduct online obituary searches for individual grantees iDoc Market Access Access to iDoc Market for Williams County, North Dakota will be required for the proof-of-concept. The selected freelancer will be responsible for creating an iDoc Market account for Williams County. I will reimburse reasonable account/subscription costs required for the test. Please do not purchase an iDoc Market subscription until instructed to do so. 1. Scrape iDoc Records For every document within the instrument types listed below, collect: Instrument Type Reception/Document Number Recorded Date Grantor(s) Grantee(s) Township Range Section Grantee mailing address Instrument Type, Reception Number, Recorded Date, Grantor, Grantee, Township, Range, and Section can be obtained from the iDoc index/search-results area without opening the actual recorded document. The grantee mailing address is not available in the iDoc index. The tool will need to open and read the actual recorded document to obtain the grantee mailing address. If Township, Range, or Section is not provided in the iDoc index, enter "-" for the applicable field. All specified instrument types should be scraped and combined into one dataset. Instrument Types to Search Administrator's Deed Administrators Deed Minerals Amended Mineral Deed Amended Personal Rep Deed Amended Personal Rep Mineral Deed Amended Quit Claim Deed Amended Quit Claim Mineral Deed Assignment of Oil & Royalty Interests Assignment of Royalty Co-Personal Rep Deed Co-PR Mineral Deed Co-PR Quit Claim Deed Co-Trustees Mineral Deed Conveyance (Minerals) Decree Deed Deed and Transfer of Royalty Interest Deed of Distribution Deed of Distribution of Mineral Interest Deed of Oil & Gas Executors Deed Executors Deed Minerals Gift Deed Gift Deed Mineral Grant Deed Grant Deed Mineral Grant of Interest in Oil, Gas, Min Guardian's Deed Mineral Guardian's Deed Judgment Mineral Life Estate Deed Life Estate Mineral Deed Mineral & Royalty Deed Mineral & Royalty Conveyance Mineral Conveyance Mineral Deed Mineral Deed and Conveyance Mineral Deed and Royalty Assignment Mineral Quitclaim Deed Mineral Right & Royalty Transfer Mineral Trust Transfer Deed Mineral Warranty Deed Oil and Gas Deed Order and Determination of Heirs Pers. Repr. Deed of Dist/Assn of Royalty Personal Representative Assign Royalty Personal Representative Royalty Deed Personal Representative Deed Personal Representatives Mineral Deed Personal Representatives Quit Claim Deed Quitclaim Deed Quitclaim Royalty Deed Royalty Deed Sheriff's Deed Special Quit Claim Deed Special Warranty Deed Trustee's Deed Trustees Deed and Assignment Trustee's Deed Mineral Trustee's Deed of Distribution Trustee's Quit Claim Deed Trustees Quitclaim Mineral Deed Trustee's Royalty Deed Warranty Deed Warranty Deed/Minerals 2. Organize and Search Records by Person After all specified instrument types have been scraped, I want the data organized so that I can search a person's name and review all documents in which that person appears as either a grantor or grantee. For each document associated with the person, I want to be able to see: Name as indexed in iDoc Role: Grantor or Grantee Instrument Type Reception/Document Number Recorded Date Township Range Section Grantee mailing address, when applicable Documents should be sortable by recorded date so that a person's appearances in the county records can be reviewed chronologically. Name searching is very important. The same person's name may be indexed differently in different documents. The tool should account for reasonable variations in how a person's name may be indexed, including: First name / last name Last name / first name Middle name or middle initial Punctuation or formatting differences For example, a search for Barbara J. Smith may need to identify records indexed as: SMITH BARBARA J BARBARA J SMITH SMITH, BARBARA J. BARBARA JEAN SMITH Names should be preserved as they actually appear in the iDoc index. The tool should not automatically assume that two similar names are the same person. The purpose is to make potentially relevant records easy to locate and review rather than having the software make a final determination about a person's identity. 3. Obituary Search For every individual person identified as a grantee in the scraped documents, conduct an online obituary search. The obituary search should use: Grantee's full name City and state from the grantee mailing address shown in the recorded document The city and state are intended to help narrow the search and distinguish between people with the same or similar names. The full street address does not need to be used for the obituary search. If a potential obituary is found, include a column in the final dataset with a link to the obituary. If no obituary is found, enter "-". Obituary searches should only be performed for individual people. Entities such as companies, trusts, banks, and government entities should be excluded. The obituary search does not need to conclusively determine whether the obituary belongs to the grantee. I will manually review potential obituary matches. Project Milestones Milestone 1 – Demonstrate iDoc Scraping This will be a paid proof-of-concept. Demonstrate that the tool can: Search Williams County iDoc by instrument type Scrape records from the iDoc index/search results Capture the required index information Open the actual recorded documents Extract grantee mailing addresses from the documents The purpose of this milestone is to demonstrate that the proposed approach works with the Williams County iDoc system before proceeding with the full project. Milestone 2 – Demonstrate Person Search and Obituary Search Demonstrate that the tool can: Search the combined dataset by a person's name Locate records where the person appears as either grantor or grantee Account for reasonable variations in how names are indexed Display the person's records chronologically Conduct automated obituary searches for individual grantees using their full name and city/state Return potential obituary links in the dataset Milestone 3 – Demonstrate Multiple Instrument Types Demonstrate that the scraper can successfully search and scrape multiple different instrument/document types listed above and combine the results into a single usable dataset. Final Deliverable Final output should be CSV/Excel and include the scraped records, grantee mailing addresses, and potential obituary links. Please explain your experience with large-scale web scraping, browser automation, OCR/document extraction, name searching/matching, and automated online research.