We want to do this in a consulting / facilitators / builders format in which we work with the facilitator / consultant / trainer for 3-6 hours a week for 3-6 months in order to help us collaboratively create various agents for our private equity business. We DO NOT want someone to exclusively build by itself (unless requested) but rather help us build together. The only billed time will be the time spent on the video call with our team, unless specifically approved otherwise. We want to be able to create a screen scrape tool to average certain cost items of specific real estate projects We also want to compare legal documents vs term sheets and excel spreadsheets Data sources • Company databases (SQL, flat files, Excel exports) - Dropbox all our files are in drop box • Extensive web scraping for competitor benchmarks and investment-market signals If you have ideas for safely adding external financial APIs later, let me know, but the two feeds above are mandatory. Tech freedom I haven’t settled on Claude, OpenAI, or any other foundation model yet. I’m open to your guidance on which stack best balances accuracy, cost, and future scalability, provided everything runs on a maintainable cloud or container platform. Core deliverables • build agents with company team to address various use cases (screen scrap competitor data, compare legal to documents to excel files, find out property values from online sources and put them in excel spreadsheet in specific format) Proven work in finance automation or similar regulated domains will weigh heavily in the selection. When you reply, please share one or two live examples of AI systems you built that involved both database integration and large-scale scraping, noting the model family you chose and why.