MatLogica | MatLogica AADC Quantitative Development Toolkit

MatLogica AADC Quantitative Development Toolkit

Complete Quantitative Development Toolkit

Proprietary JIT graph compiler that delivers 6-1000x speedups and automatic derivatives with <1% code changes. Focus on your models—AADC handles performance and AAD automatically.

Get All Your Greeks for Free—Actually Faster Than Just Computing Price

AADC's revolutionary adjoint factor <1 means computing your function plus all derivatives faster than your original code computes just the function alone

The AADC Toolkit

The comprehensive solution for compute-intensive calculations and automatic differentiation

The Two-Fold Challenge Institutions Face:

1. Ever-Increasing Simulations

  • New regulatory requirements (FRTB, SA-CCR)
  • More scenarios and stress tests
  • Real-time risk demands
  • Larger portfolios to analyze

2. Need for AAD (Greeks)

  • Complete sensitivities required
  • XVA with all Greeks
  • Model risk management
  • Advanced risk metrics

All complicated by decades-old legacy systems that can't easily be rewritten

AADC: The Unified Solution

AADC Toolkit addresses both challenges simultaneously while simplifying legacy integration. It's a comprehensive suite of tools featuring MatLogica's patented JIT graph compiler, powered by Code Generation AAD™ technology, which automatically compiles numerically intensive functions into highly optimized machine code with full derivatives.

Simulation Acceleration

6-100x faster Monte Carlo, scenarios, and what-if analysis

Automatic AAD

Complete Greeks and sensitivities with adjoint factor less than 1

Legacy Simplification

Minimal code changes, no complete rewrite required

Six Core Components for Complete Control

Everything you need for production-grade quantitative development

🚀

AADC Engine

Proprietary JIT Compiler

  • Code Generation AAD™ technology
  • Near-instantaneous compilation (milliseconds)
  • Automatic vectorization (AVX2/AVX512)
  • Automatic multi-threading
  • Adjoint factor <1 achievement
🔧

Integration Scripts

Automated Migration

  • Search & replace: double → idouble
  • Generic type conversions
  • Automated migration utilities
  • Minimal refactoring needed
  • Typically <1% code changes
🔍

Debugging Toolkit

Reverse Debugger & Diagnostics

  • Compare kernel vs original values
  • Monitor adjoint propagation
  • Numerical discrepancy detection
  • Bump & revalue troubleshooting
  • Intermediate variable tracking
🌿

Branch Manager

Handle Complex Control Flow

  • Static branches: Hard-coded into kernel
  • Dynamic branches: bool → ibool conversion
  • Smooth discontinuous payoffs
  • Critical for barriers & autocallables
  • Automated conversion reports
⚙️

AIFT Solver Support

Calibration Without Refactoring

  • Automated Implicit Function Theorem
  • Derivatives through optimization
  • No code refactoring needed
  • Most downloaded Risk.net article 2022
  • Key technique for Live Risk
📚

Reference Implementations

Production-Ready Blueprints

  • Open-source on GitHub
  • Phoenix autocallables with AAD
  • XVA frameworks
  • American Monte Carlo (Longstaff-Schwartz)
  • Python DSL for rapid prototyping
MatLogica GitHub
How AADC Works

Choose Your Implementation Path

Three proven routes to take advantage of AADC

🔄

Legacy Optimization

Modernize existing code without rewriting

  • Code runs but too slow
  • Can't afford full rewrite
  • Risk of breaking models
  • Limited developer resources
THAT'S ME
🚀

Greenfield Build

Develop custom capabilities 5x faster

  • High vendor licensing costs
  • Slow custom development
  • Need proprietary models
  • Time-to-market pressure
THAT'S ME
⚡

Accelerate Open-Source

Supercharge QuantLib/ORE by 6-100x

  • Open-source too slow
  • Need regulatory acceptance
  • Want proven algorithms
  • Competitive disadvantage
THAT'S ME

No, I just want to save on cloud

See Cloud Cost Optimization →

No, I just want fast Python

nav-tech-python-accelerator

Industry Proven

Explore independent benchmarks, whitepapers, and validation studies demonstrating MatLogica AADC's performance across diverse workloads. From Intel-backed research showing 1770x speedups to open-source comparisons with JAX and PyTorch, see the evidence.

See how our clients transformed their quant libraries

What You'll Learn
  • Overnight Batch - from 8 hours to 2 hours
  • Developer Productivity Increase
  • From 0 to prod in 1 year
View Case Studies

What Experts Say

"It's hard to develop well-performing models in C++. I've been very impressed by performance gains up to two orders of magnitude obtained by MatLogica after some integration work on QuantLib, and by the fact that the same work also enabled AAD; especially considering that the library contains hundreds of thousands of lines of code developed over more than 20 years."

Luigi Ballabio
Co-founder and administrator of QuantLib

"MatLogica's product changes the paradigm for quantitative software development eliminating the need to invest in optimizations. Quants can now focus on the models, and performance will be taken care of by MatLogica's JIT compiler."

Paul A. Bilokon
CEO, Thalesians Ltd; Visiting Professor, Imperial College London

"Matlogica offers a graph-based framework in C++, optimized for finance applications on modern hardware aware and complete with state-of-the-art AAD and proprietary on-the-fly compilation. Quants will find that Matlogica seamlessly fits with their libraries and effortlessly accelerates pricing and risk by impressive amounts."

Antoine Savine
Author of "Modern Computational Finance"

"The approach MatLogica takes to acceleration is novel in both its easy-to-use programming interface and high performance it achieves out-of-the-box. Straight-forward and minimal code changes, to make use of the libraries, offer leaps in performance for both xVA Pricing and Greeks Calculations."

Mahesh Bhat
Principal Engineer, Intel

Who Uses AADC Toolkit

Quants & Developers

Focus on models not performance. 3-5x faster development. Eliminate manual SIMD, threading, and AAD coding.

Front Office

Real-time pricing and Greeks for competitive advantage. Ultrafast RFQ response. Complex exotics priced in seconds.

Risk Management

Real-time VaR and regulatory compliance. 50-150x XVA acceleration. FRTB and SA-CCR deadlines met with ease.

Technology Leaders

Strategic modernization without risky rewrites. 50%+ cloud cost reduction. Vendor independence. 20x ROI by year 2.

Front Office Solutions
Tech Solutions

Technical Deep Dive

How AADC Works

In-depth explanation of the compiler technology and optimization techniques that make MatLogica AADC significantly faster than traditional AAD approaches.

Business Impact
  • "20-50x faster than original analytics
  • Potential for ~100x cloud cost reduction
  • Multi-core scaling without code modification
  • Secure deployment - models and data stay on premises
  • No latency - kernels ready at market open"
Technical Details

Adjoint Factor <1

Revolutionary achievement: compute function + all derivatives faster than original function alone. Breaks the theoretical 4x minimum with adjoint factor <1.

Business Impact
  • Get all Greeks for 'free' - actually negative cost
  • No trade-off between speed and derivatives
  • Makes AAD adoption compelling everywhere
Read Research

Mix-Mode Execution

Seamless integration of C++, Python, and C# code in a single computational graph. Record in one language, execute from another, or mix in single valuation graph.

Business Impact
  • Use best language for each component
  • No performance loss at language boundaries
  • Incremental modernization path
  • Team flexibility (Python + C++ developers)

Supported Languages & Platforms

C++
Full support
Python
Full support
C#
Supported
Java
In progress

Hardware: Intel/AMD CPUs with AVX2/AVX512 • GPU support in roadmap

Flexible Licensing Options

Enterprise License

Organization-Wide Deployment

  • All languages and features
  • Full technical support
  • Integration consulting available
  • Production use cases
  • Development licenses at reduced cost

Contact us for pricing based on your organization size and use cases

Desk/Machine-Bound License

Individual Development

  • Licensed per machine
  • Prototyping and custom models
  • Individual quant development
  • Bottom-up adoption strategy
  • Try before enterprise commitment

Perfect for evaluation and individual use cases

Free Options: Demo licenses for evaluation • Academic licenses for educational institutions

Get Started with AADC Toolkit

See how AADC can transform your quantitative development

Schedule Consultation

Explore Related Solutions

Industry-Leading Recognition and Validation
Business Impact
  • Chartis Category Leader 2022-2024 (4 categories)
  • RiskTech100 2024 featured company
  • Asia Risk Technology Newcomer 2024
  • QuantTech50 top quartile ranking
View All Awards
Python Accelerator
Business Impact
  • Makes Python production-ready for Live Risk
  • Eliminates Python-to-C++ rewrite cycle
  • Enables Python-first development strategy
  • Performance exceeds hand-optimized C++
Python Accelerator
Integration Process
View Technical Details