Keystone

About

The K.ATS team comprises ~50 computer scientists, data scientists, engineers, and technologists supported by a network of dozens of leading computer science and AI experts from industry and academia across the globe.

‍

K.ATS has the on-premises and cloud-based infrastructure and capabilities to ingest, parse, and analyze a wide variety of highly confidential big data (>100TB) and source code from the world’s largest tech companies. Our cleanroom and other proprietary tools offer a secure end-to-end data transport, storage, analysis, and disposal solution.

‍

The K.ATS team works with clients on the world’s most consequential strategic, legal, and regulatory matters impacting technology. Prior K.ATS project work includes:

Source Code Measurement and Review

The review, analysis, and measurement of complex codebases to estimate the value of software technology for IP and tax matters
Code review and forensic analyses to determine patent, copyright, trade secret, data breach, and related IP theft

‍

AI/ML Training, Auditing, and Valuation

Analyzing the impact and value that specific training data has on AI/ML model performance (including Generative AI models)
Training and deploying large language models (LLMs) with guardrails and monitoring frameworks
Assessing how “black box” AI/ML models work and their capability for identification (e.g., for BIPA matters)
Conducting algorithmic audits to assess large-scale AI/ML systems (e.g., recommender systems) for fairness and bias concerns

‍

Cybersecurity, Privacy, and Regulation‍

Architecting and implementing privacy-enhancing technologies (PETs) to safeguard sensitive user data for regulatory compliance matters
Analyzing network traffic communications to determine information sent (e.g., via pixels) and the “architecture” of those communications between client/servers for VPPA, CPPA, and wiretap-related matters
Analyzing data pipelines set up to scrape web content and bot traffic on social media platforms for CFAA and Generative AI-related matters

‍

Tech Strategy and Architecture

Software and hardware technology teardowns and architecture/integration assessments (e.g., distributed systems, semiconductors, chips) for IP and antitrust investigations
Developing data, platform, and system strategy roadmaps and tech diligence reports for M&A, including assessments of quality, scalability, security, and debt in systems

‍

Capabilities

Source Code Value Measurement and Analytics

Informing granular understanding of code features and value (e.g., lines of code, dependencies, code carryover, technical debt)

Source Code Review and Digital Forensics

Applied to large-scale, real-world distributed systems, like ad auction infrastructure, recommender systems, and information retrieval systems; interception of mobile app networking communication and reverse engineering

Machine Learning Development and Analytics

Including model and pipeline design, development, and deployment, and algorithm auditing to explain and assess measures of bias, fairness, interpretability, data value, and privacy

LLM Training and Data Value

Conducting experiments quantifying training data’s impact/value on LLM performance/output (e.g., training LLMs, assessing datapoint marginal value); analyzing large-scale telemetry data to quantify user behavior patterns of LLM interactions and relationship with outputs

Cybersecurity and Risk Assessments

Including design, development, deployment, and advisory on responsible ML systems, privacy-enhancing technologies like differential privacy and multi-party computation, risk assessment for regulatory compliance with DSA, DMA, GDPR, etc.

Technology Architecture Development

Applied to large-scale system design, platform architecture, and implementation advisory and development including AI/ML systems and Generative AI-based apps

Keystone Advanced Technology Services

Source Code Measurement and Review

AI/ML Training, Auditing, and Valuation

Cybersecurity, Privacy, and Regulation‍

Tech Strategy and Architecture

Source Code Value Measurement and Analytics

Informing granular understanding of code features and value (e.g., lines of code, dependencies, code carryover, technical debt)

Source Code Review and Digital Forensics

Applied to large-scale, real-world distributed systems, like ad auction infrastructure, recommender systems, and information retrieval systems; interception of mobile app networking communication and reverse engineering

Machine Learning Development and Analytics

Including model and pipeline design, development, and deployment, and algorithm auditing to explain and assess measures of bias, fairness, interpretability, data value, and privacy

LLM Training and Data Value

Conducting experiments quantifying training data’s impact/value on LLM performance/output (e.g., training LLMs, assessing datapoint marginal value); analyzing large-scale telemetry data to quantify user behavior patterns of LLM interactions and relationship with outputs

Cybersecurity and Risk Assessments

Including design, development, deployment, and advisory on responsible ML systems, privacy-enhancing technologies like differential privacy and multi-party computation, risk assessment for regulatory compliance with DSA, DMA, GDPR, etc.

Technology Architecture Development

Applied to large-scale system design, platform architecture, and implementation advisory and development including AI/ML systems and Generative AI-based apps

Rohit Chatterjee

Junsu Choi

Stefan Hunt

Anna Lischke

Max Whitmore

News & Publications

Revolutionizing Clinical Trials with AI

Keystone Unveils New Branding Reflecting Expansion of AI Tech Services

You Are What You Eat: Nurturing Data Markets to Sustain Healthy Generative AI Innovation

How can we help you?