NMTC Application Builder¶

The open-source intelligence platform for competitive CDFI Fund applications.

New Markets Tax Credit allocation applications are decided on margin. Two CDEs with comparable missions and track records can receive divergent outcomes because one pipeline had 82% deep-distress concentration and the other had 67%. One had projects in 9 states; the other had 4. NMTC Application Builder exists to close that information gap — giving every CDE the same benchmarking intelligence that previously required expensive consultants or years of pattern-recognition experience.

The library reads your project pipeline, enriches each project with NMTC eligibility and distress-level data, and scores your application against the CDFI Fund's published CY 2024-2025 Review Process criteria — Business Strategy (0–50), Community Outcomes (0–50), and Priority Points (0–10). It identifies exactly which section minimums you are and aren't meeting, generates prioritized recommendations with CDFI Fund citations, and produces a complete Word/Excel/PDF/Markdown application package. All in Python, all open source.

What it does¶

Pipeline Analysis

Enriches each project with NMTC eligibility (census tract lookup via nmtc-mapper) and deal economics (NMTC credit calculation via nmtc-calc). Produces distress concentration, geographic diversity, sector mix, and impact metrics in one call.

Win Alignment Scoring

Scores your application against the CDFI Fund's published CY 2024-2025 criteria: Business Strategy (0–50), Community Outcomes (0–50), and Priority Points (0–10 bonus). Returns an aggregate base score (0–100) and tier classification: Not Qualified / Highly Qualified / Top Tier. This is not a win probability — see the methodology note below.

Actionable Recommendations

Generates specific, quantified recommendations ranked by priority (critical / high / medium). Each recommendation includes a numeric improvement estimate, not just generic advice.

Pipeline Optimization

Uses greedy construction + swap-based local search to find the subset of your pipeline that maximizes alignment score subject to QEI budget, project count, state diversity, and sector constraints.

Document Generation

Renders a complete NMTC application package — CDFI Fund section structure (A through E), all required data tables, and supporting visualizations — in Word, Excel, PDF, and Markdown.

60-second quickstart¶

This block runs on the shipped fictional CDE

CDEProfile.sample() is Riverbend Community Capital CDE, LLC — a CDE that does not exist. Nothing in the package refuses this path (the refusal is on CDEProfile.from_yaml() and the Streamlit upload, not on sample() or generate()), so the four files below are filing-shaped documents for a fictional applicant. Run it to see the tool work; replace CDEProfile.sample() with CDEProfile.from_yaml("cde_profile.yaml") before anything you would file. nmtcapp init <dir> scaffolds that file.

from nmtcapp.core.application import Application
from nmtcapp.core.cde import CDEProfile
from nmtcapp.core.pipeline import Pipeline

cde = CDEProfile.sample()
pipeline = Pipeline.sample(n=20)
app = Application(cde=cde, requested_allocation=65_000_000)
app.add_pipeline(pipeline)
paths = app.generate("./drafts/")

That five-line block runs all intelligence analyses and writes a complete application package (Word, Excel, PDF, Markdown) into ./drafts/. It prints nothing: generate() logs at INFO and writes no output to the terminal. For the summary on screen, call app.analyze().summary().

Get started with Installation Read the Quickstart View on GitHub

Methodology Disclosure

The score produced by score_win_probability() measures alignment with the CDFI Fund's published CY 2024-2025 evaluation criteria — it is not a probability of receiving an allocation. The CDFI Fund's actual scoring rubric is proprietary; sub-score weights are this tool's best-effort interpretation of the published guidance. A high score reflects strong self-assessed positioning; it does not guarantee funding. Always have a qualified CDFI/NMTC practitioner review your application before submission.