Built This Week/Guests

Guests

Eldar Sadikov

Eldar Sadikov founded Field Materials AI, which automates material and equipment procurement for commercial and civil contractors with AI agents that read procurement documents and plug into ERPs.

Eldar Sadikov is the founder of Field Materials AI, an AI platform that automates material and equipment procurement for commercial and civil contractors. Its agents read supplier quotes, delivery tickets and invoices, then do the data entry and checks that purchasing and accounts payable staff used to do by hand. He joined us in Episode 33 (February 27, 2026).

Eldar went to Stanford for grad school and founded his first company, an AI company for retail, in 2011. He started Field Materials in January 2022, the same year ChatGPT came out, and launched the product in 2023. He said the company grew about 7x the year before last and 3x last year.

What Field Materials AI does

About 80% of Field Materials' customers are commercial specialty trade subcontractors: electrical, mechanical (HVAC and plumbing), concrete and drywall contractors building hospitals, data centers and office buildings. They buy thousands of materials per project. Today, Eldar said, most of that runs on emails, spreadsheets and someone typing every line item into accounting software that is often still on-premise.

Field Materials reads the documents, automates the data entry and verification, and keeps a human in the loop. Tight integrations with the contractor's ERP let it automate whole workflows, not just extraction.

The demo

At 13:05, Eldar showed the agent lineup on screen. Field Materials builds a dedicated agent for each document type: switchgear quotes, HVAC quotes, trucking tickets, asphalt tickets and so on. Each one knows how to read its document, extract the line items, and verify and match them. There are now more than 20 of these agents, fine-tuned on thousands of customer documents. Next up are agents that carry out ERP and procurement steps themselves, not just read documents.

Because we're using these foundational models, when these foundational models get better every single quarter, every single month, our fundamental ability to read, interpret, and verify the documents is also going up.

— Eldar Sadikov, Episode 33

Key ideas from the conversation

  • Double the volume, same staff (10:41). A contractor going from $100 million to $200 million in projects can handle it without doubling purchasing and AP headcount. Some customers buy it because an AP person is retiring and can't be replaced.
  • Stop margin bleed on materials (11:20). Waste, double orders and paying retail instead of the negotiated rate eat margin. Automated verification makes sure contractors pay the correct price and cuts rogue spending.
  • Faster payment cycles (12:11). Invoices that took 30 to 45 days to approve now take a few days, and paying early can earn discounts.
  • The data center boom is forcing change (14:53). Construction has spent little on IT for decades. Aggressive data center schedules are pushing contractors off pen and paper and manual entry.
  • Start from the use case (19:21). On SAP customers questioning AI value, Eldar said bolting AI onto old software just to check a box produces a gimmick. Quality AI tools take many iterations.
I believe we're we're sort of entering this golden era for construction technology, and AI is a big element of this.

— Eldar Sadikov, Episode 33

The key is to make it useful.

— Eldar Sadikov, Episode 33

Eldar's view lines up with what Tristan Wilson of Edgevanta told us about estimating in Episode 16: agents take the data entry, and a person still reviews the result. More on both on our AI in construction page.

What we built for the episode

Sam opened with a BOM price-shock scenario tool based on a possible future feature he had discussed with Eldar. You upload a bill of materials and change prices line by line: a 10% jump in concrete showed a 6% risk of overrun. Shock presets model a tariff shock (steel and rebar, up to 9% overrun risk, with a suggestion to lock steel pricing via forward contract), a port delay (1% risk, mostly plumbing) and market relief. Eldar said it resembled Field Materials' pricing intelligence and BOM modules, and that the variable-driven scenarios could be good additions to the platform.

FAQ

Who is Eldar Sadikov?

Eldar Sadikov is the founder of Field Materials AI. He went to Stanford for grad school and founded his first company, an AI company for retail, in 2011. He started Field Materials in January 2022.

What does Field Materials AI do?

Field Materials AI automates material and equipment procurement for commercial and civil contractors. Its AI agents read supplier quotes, delivery tickets and invoices, automate data entry and verification with a human in the loop, and integrate with the contractor's ERP.

Who uses Field Materials AI?

About 80% of its customers are commercial specialty trade subcontractors, such as electrical, mechanical, concrete and drywall contractors building hospitals, data centers and office buildings.

How much time can AI save on construction invoices?

Eldar said invoices that took 30 to 45 days to approve can be processed in a few days, which can also earn contractors early-payment discounts.