Tax Law. Software Engineering. Artificial Intelligence.

Crimson Tree Software is a pioneering software development team specializing in US partnership taxation.

Over two decades, our products have served CPA firms, real estate investment firms, hedge funds, private equity investment funds, and corporate joint ventures.

Since 2013, we have focused on helping large CPA practices solve complex issues relating to transaction analysis and partnership tax compliance.

We build intuitive, human-centric solutions that elevate quality, increase productivity, and reduce delivery costs.

Our team unites experienced software engineers, accounting and finance professionals, and cloud-computing specialists.

We take pride in our deep business knowledge, empathy for customers, openness to new ideas, and commitments to active listening and critical thinking.

In 2023, systems invented and built by Crimson Tree Software service over $2 billion in recurring annual fees.

Core Competencies

  • The internal revenue code, regulations, and industry practices relating to US partnership taxation
  • Cloud-native technologies including expertise with PaaS and SaaS architectures
  • Implementation of strong data privacy and cyber security defensive mechanisms
  • Iterative product development using Agile practices, DevOps, CI/CD, and automated testing
  • Invention of high-performance, parallelized, in-memory computations in big data applications
  • Multi-strategy AI approaches, including heuristic algorithms, applied statistics, machine learning models, and generative AI
  • Open data architectures facilitating integration with upstream and downstream systems
  • Human-centric user experience design

Partnership Tax AI

To be released in closed beta in Q12024. Please contact us if you are interested in participating.

Development of a new and practical artificial intelligence system for comprehension of K-1s, K-1 footnotes, and other associated tax documents.

Integrating multiple strategies including generative AI, Bayesian inference, pattern recognition, self-programming, heuristic algorithms, and traditional software automation, this system understands and extracts data from complex documents with unprecedented speed and accuracy.

In the simplest use, users can pass in a single PDF document, or set of documents, and receive a data package with annotated PDF pages and Excel files containing all extracted data organized in regularly formatted tabs.

In interactive use, users can search, sort, filter, slice, and dice data across their document repository, and can query their vectorized text through a ChatGPT interface.

Our Tax AI can comprehend hundreds of semantic concepts and data types, including but not limited to:

  • The face of the K-1, with self-programming to account for document template variations
  • Overflow statements and other form face-related detail footnotes
  • Other critical footnotes including DRD, ECI, UBTI, FDAP, PFIC, Installment Sales, Passive Activity, and QBI / 199A
  • State source income matrices
  • State K-1s including complex forms such as the CA, NY, and PA K-1s
  • Classification of a variety of additional forms including 926, 8621, 8865, 8938, and 8886

Next-Gen Partnership Tax Compliance System

Development of a comprehensive partnership tax compliance system using a cloud-first strategy and modern system architecture.

The sale of this system to an international accounting firm was completed in 2022 and by 2023 yielded a nine-figure return on investment.

This system comprehensively addresses the needs of a large partnership tax practice, including the following key features:

  • GAAP to Tax adjustments, allocations, and tiering for partnerships and tiered organizational structures of any size and complexity
  • Primary calculation pipeline for federal, state, and international tax reporting
  • Secondary calculation pipelines for complex Section 704(c) allocations found in real estate partnerships and hedge funds
  • Rapid calculation turnaround for even the largest partnership structures
  • Real-time, in-memory, self-service analytics and reporting functions
  • All inputs and outputs exportable in XLSX and compressed CSV formats
  • Simple integration with Excel, Alteryx, PowerBI, and other end user tools
  • Direct generation of high-quality K-1s, footnotes, K-3s, and K-2
  • Systems integration with widely used tax return filing systems such as GoSystem and CCH
  • Append-only data model with full revision history, tracing capabilities, and the ability to reverse all data changes
  • Instantaneous copying of datasets, with copies made from any point in their revision history
  • Configurable data ingestion and transformation functions for client data sources such as trial balances, balance sheets, and income statements

Partnership Modeling Platform

Development of a deeply technical partnership transaction modeling platform in collaboration with a Big 4 National Office M&A practice.

Among other projects, in 2019 the system was used to propose and model the Up-C conversion of a prominent publicly traded entity. Upon conversion, the entity enjoyed a significant increase in market value. Other businesses with similar structures followed.

Core functions include:

  • Tracking over time of partners’ varying 704(b) interests, section 704(c) differences, and capital accounts. Automation of internal accounting relating to transfers of interest and redemptions.
  • Tracking of section 704(b) book basis and tax basis of partnership property, including depreciable and amortizable property.
  • Section 704(b) revaluations of partnership property.
  • Section 704(c) depreciation and gain/loss allocations under traditional, remedial, and curative methods. Other mixed method approaches in which 704(c) methods differ by revaluation layer.
  • Section 752 debt analysis and allocations allowing for changes in debt fact patterns.
  • Calculation of hypothetical liquidations including gain/loss calculations at the asset and partner level.
  • Section 751(a) gain recharacterization analysis.
  • Section 751(b) gain analysis on distributions that result in deemed sales or exchanges.
  • Section 731 gain analysis on distributions.
  • Section 732 gain analysis on distributions of partnership property.
  • Section 743(b) adjustment calculations, allocation to property, and ongoing recovery, including both two-way and one-way adjustments.
  • Section 734(b) adjustment calculations, allocation to property, and ongoing recovery.
  • Allocation of miscellaneous other 704(b) book and tax items to partners.

Automated Extraction of Structured Data from Tax Documents

Development of several machine learning and heuristic-AI based applications that extract structured data from tax forms and complex footnotes.

These projects succeeded before the advent of modern Large Language Models and Generative AI.

A difficult area in which we made significant progress was the classification and extraction of data from complex footnotes that trigger reporting requirements (e.g., forms 926, 8865, 8886, and 8621).

By 2020, these systems were trained to recognize and extract data from 44 distinct classes of footnote.

The core libraries united open-source software, closed-source commercial software, and custom developed software to produce a unified and royalty-free system for extraction of data from tax documents.

Examples of deployed applications include:

  • A fund-of-funds focused system for ingestion and aggregation of K-1s, overflow schedules, and footnotes. Source documents of varied quality, ranging from high quality searchable PDFs to low quality scanned images. Deployed for use both by line of business staff and a global delivery center. Eliminated 30,000 labor hours in the first year of use.
  • A private-equity focused system to ingest, review, and enhance K-1s in scenarios where the funds receive K-1s from hundreds or thousands of underlying entities.
  • A system for ingestion and general-purpose query of field data from over 2,000 form types (inclusive of variations).

Tax Adjustments to Financial Transactions

Development of a comprehensive system for calculating US tax adjustments to financial transactions.

Clients include prominent hedge funds and regulated investment companies. Many individual client portfolios have more than 1 million transactions subject to analysis each tax year.

The core tax adjustment calculations include:

  • Wash sale adjustments
  • Constructive sale adjustments
  • Tax straddle adjustments
  • Basic tax straddles
  • Identified tax straddles
  • Mixed straddles
  • Section 1256 MTM adjustments
  • Section 1256 reclassification adjustments
  • Section 1296 PFIC MTM adjustments
  • Debt adjustments under Sections 1276, 1278, 171, and 1283
  • Section 988 reclassification adjustments
  • Section 988 bond bifurcation
  • Section 988(a)(1)(b) election analysis
  • Section 1233 short position analysis
  • PTP adjustments for RICs
  • Unsettled shorts analysis
  • Substantially worthless security adjustments
  • Return of capital adjustments
  • Redemptions in kind
  • Acquisitions in kind
  • Section 1061 analysis (Long Term Capital analysis for incentive recipients)
  • Qualified dividend analysis
  • Dividends received deductions
  • Foreign tax credits
  • Short dividends Section 263(h) analysis (capitalized short dividends)
  • REIT distributions analysis
  • RIC distributions analysis
  • Section 199A analysis

Partnership Tiering and Allocation Engine

In 2011, we began experimenting with novel software designs incorporating in-memory analysis and ubiquitous parallel processing.

This work yielded a new, high-performance system of algorithms to compute US partnership tax allocations pursuant to IRC Subchapter K.

In 2013, a Big 4 accounting firm purchased this partnership allocation engine and portfolio of related inventions.

Our team was then engaged in a successful multi-year project that resulted in a comprehensive, “GAAP to Tax” partnership tax compliance system.

Our contributions include:

  • All federal and state tax calculation logic
  • Design of the system’s tax-domain conceptual model and its overall data model
  • All system middleware connecting the calculation logic and cloud-computing processing and data infrastructure
  • End-user workflow for book to tax and other key tax processes
  • Federal and state form production functions
  • Footnote production functions including free-form HTML text formatting and dynamic compilation of user-specified functions

Today, our systems remain at the core of the firm’s partnership tax compliance practice, with over 6,000 daily users during peak periods.

Other Projects

  • Creation of a machine-learning based system for the ingestion and mapping of client general ledger data to a universal tax-sensitized system of classification.
  • Creation of standardized web-based corporate workpapers for ingestion of PBC data and the automation of book to tax workflow processes.
  • Creation of a reusable MACRS and ADS depreciation calculation library along with a comprehensive, 25,000+ asset test case used to verify its correctness.
  • Relating to the calculation of GILTI, or Global Intangible Low Tax Income, creation of a high-performance calculator to hypothetically depreciate non-US assets as though they were under the ADS, alternative depreciation system, in order to calculate hypothetical Qualified Business Asset Investment income.
  • Creation of a library for deep validation of E-File XML to match IRS and state revenue requirements. Scripted rules allow for deep validation of XML, including both syntax validation of individual fields and semantic validation of fields in relation to one another.
  • Creation of diagnostic analysis and visualizations relating to the misreporting of trading data by brokers and other nominees holding certain securities in street name.
  • Numerous optimization projects relating to performance bottleneck identification in existing systems followed by creation, testing, and integration of new, high-performance, parallelized algorithms and dataflows.

Locations

Arlington, VA
Maple Lawn, MD
Portsmouth, NH
Dallas, TX