PermitAI is developing an enriched, large-scale database that provides seamless access to thousands of historical environmental reviews and permitting documents. Data collection was initially focused NEPA documents, including categorical exclusions, environmental assessments, environmental impact statements, and other associated documents. This ongoing effort to create a centralized and standardized repository of NEPA documents from across the federal government is known as NEPA Text Corpus (NEPATEC). 

Since its inception, the NEPATEC database was designed to be extensible to other permitting review documents, which now includes adjudication records and other interconnected permitting artifacts. As PermitAI expands, the data lakehouse is designed to integrate information across regulatory processes and decision stages, while beginning to incorporate select state-level permitting data. PermitAI is starting with geothermal energy and critical mineral permitting use cases in Alaska and Nevada, enabling a more complete, end-to-end view of the permitting lifecycle and supporting more transparent, efficient, and data-driven decision-making.

NEPATEC

Released to the public in June 2024, NEPA Text Corpus (NEPATEC) 1.0 is the first iteration of PermitAI’s expansive database of federal agency NEPA documents, consisting of more than 28,000 documents from nearly 3,000 projects across more than 100 agencies. The public release of NEPATEC2.0 delivered an expanded corpus of NEPA documents, consisting of more than 120,000 documents from 60,000 projects prepared by more than 60 different agencies. By using large language models to extract metadata modeled to align with Council of Environmental Quality (CEQ)’s NEPA and Permitting Data and Technology Standard, NEPATEC2.0 promotes consistency in environmental reviews and supports the ongoing effort to modernize permitting technologies by facilitating more transparent, efficient, and data-driven decision-making.  Targeted for a Q3 2026 release, NEPATEC3.0 will add an estimated 60,000 new documents to the database along with Geographic Information System (GIS) elements and refined textual metadata.

Beyond NEPA: PermitTEC v0.1

PermitTEC v0.1 expands PermitAI’s data lakehouse to include the legal side of permitting, not just environmental review documents. This public dataset provides clean, standardized, machine‑readable metadata on federal court cases involving NEPA and related environmental laws. It complements NEPATEC by showing how environmental review decisions are challenged and resolved in court, linking cases to projects when possible. By combining litigation data with review documents, PermitTEC gives a more comprehensive picture of the permitting process and helps reveal how project features, environmental factors, and procedural choices relate to legal outcomes. Built to match CEQ data standards, it supports more transparent and data‑driven decision‑making.

Data Innovations

PermitAI uses advanced data engineering and data science to build the data lakehouse that that serves as the foundation for NEPATEC and several PermitAI tools. 

The team works closely with federal agencies and CEQ to gather NEPA and permitting documents, including automatically pulling environmental impact statements from the EPA’s CDX API. All documents are stored in a secure cloud system designed for fast search and long‑term preservation.

PermitAI automates many routine data tasks to improve efficiency and protect data quality. Beyond storing documents, it is also developing standardized, enriched metadata using large language models. This metadata captures key details about projects, processes, documents, public comments, and geographic information, creating a stronger foundation for analysis.

PermitAI Data Lakehouse Team

• Dan Nally (Data Thrust Lead)
• Tim Vega (Data Engineering Lead)
• Kaustav Bhattacharjee (Data Science Lead)
• Anurag Acharya (Data Thrust PM)
• Jim Jackson (Domain Lead)
• Beau Morton (Data Engineering Lead)
• Rounak Meyur (Data Science Lead)
• Aaron Moreno
• Alex Buchko
• James Bandy
• Kathy Nwe
• Nathan Butschli
• Matthew Raffel
• Sai Koneru
• Siddhartha Das
• Sridevi Wagle
• Ben Chauhan
• Heng Wan
• Johnny Chen
• Micah Taylor
• Rizwan Ashraf
• Paul Rigor
• Greg Wint
• Joshua Wassing
• Thomas Serrano
• William Zhang
• Michael Kieburtz
• Brian Chen
• Cole Man
• Meenu Mohankumar
• Ellyn Ayton
• Renuka Chintalapati
• Milan Jain
• Julianna Puccio
• Sam Donald