Expert Witness &
Technical Consulting

Technical expert witness, source code forensics, and technical due diligence for attorneys, investors, and businesses. 25 years of hands-on experience, court-ready expert reports filed under Daubert standards, and fast turnaround on every engagement.

What We Do

Broad technical consulting for businesses, law firms, and startups. Deep expertise and modern tools for fast, thorough results.

Software, Firmware & Kernel Engineering

AI-directed software, firmware, and kernel-driver engineering. Peter directs Claude Code Max on deep technical projects — delivering scope that used to require a team, at solo-practice pricing. See the Software Development section for details.

  1. Free 30-minute scoping call — define requirements, target hardware, and timeline
  2. Sign engagement letter with scope and rates
  3. Grant access to repos, hardware, or target systems
  4. Rapid iteration with weekly check-ins; deliverables verified on real targets
📊

Data Analysis & Business Intelligence

Financial modeling, reporting dashboards, large-scale data processing. Make sense of your data and turn it into actionable decisions.

  1. Free 30-minute scoping call — what questions need answering?
  2. Sign engagement letter; NDA if working with sensitive data
  3. Share data via secure upload or grant database access
  4. Receive analysis report with visualizations and recommendations

Infrastructure & DevOps

Linux administration, cloud architecture, CI/CD pipelines, monitoring, and security hardening. Keep your systems running smoothly.

  1. Free 30-minute scoping call — current setup and pain points
  2. Sign engagement letter; VPN/SSH access as needed
  3. Audit current infrastructure; propose improvements
  4. Implement changes with documentation and runbook
🤖

AI/ML Consulting

LLM integration, RAG systems, automation workflows. Practical AI that solves real business problems — not science projects.

  1. Free 30-minute scoping call — what problem are you trying to solve?
  2. Sign engagement letter with scope
  3. Prototype phase — prove the approach works before scaling
  4. Deliver working integration with documentation
🔍

Litigation Support & Expert Witness

Document review, source code forensics, contract analysis, and expert reports filed under Daubert standards. From terabytes of raw data to court-ready deliverables — with automated quality verification on every document.

  1. Conflict check and free 30-minute consultation with counsel
  2. Sign retention agreement and protective order if applicable
  3. Receive data — secure upload, courier drives, or server access
  4. Analysis, reports, and expert testimony as needed
📋

Technical Due Diligence

Pre-acquisition code audits, software asset valuation, architecture assessments, open source license compliance. Fast turnaround for deal timelines.

  1. Free 30-minute scoping call — deal timeline and what needs evaluating
  2. Sign NDA and engagement letter
  3. Access to data room, codebase, or technical documentation
  4. Receive written assessment with findings and risk summary

Core Practice Areas

Focused expertise across four practice areas where deep technical skills, rigorous methodology, and modern tooling matter most.

Expert Witness Services

Peter Gransee provides technical expert witness and software expert witness services for IP disputes, trade secret cases, patent litigation, and contract disputes involving software, source code, and engineering work product. Testimony, expert reports, and deposition support for federal and state courts.

Expert witness engagements include:

  • Expert reports filed under Daubert standards, with cited exhibits and reproducible methodology
  • Source code review, authorship attribution, and line-by-line similarity analysis
  • Deposition preparation and trial testimony
  • Trial exhibits, demonstrative graphics, and technical tutorials for judges and juries
  • Rebuttal reports responding to opposing expert reports

Typical cases: IP ownership disputes, trade secret misappropriation, software copyright claims, patent infringement involving source code, and contract disputes over software deliverables and acceptance criteria. When you need a technical expert witness who can explain source code to a jury and stand up to cross-examination, that's the work.

Technical Due Diligence & Software Due Diligence

Pre-acquisition technical due diligence and software due diligence for investors, M&A advisors, private equity firms, and corporate acquirers. Fast turnaround matched to deal timelines — typically one to three weeks depending on code base size and review scope.

Technical due diligence reviews cover:

  • Source code quality, architecture, and maintainability assessment
  • Open source license compliance — detecting GPL contamination, attribution requirements, and redistribution obligations
  • Security posture review — exposed secrets, dependency vulnerabilities, authentication patterns, and data handling practices
  • Scalability analysis and technical debt quantification
  • Team capability and key person risk assessment
  • Software asset valuation for deal pricing and negotiation support

Deliverable: Written technical due diligence report with executive summary, detailed findings, risk rating, and specific remediation recommendations. Suitable for sharing with deal partners, lenders, and legal counsel. Every technical due diligence engagement is backed by a confidentiality agreement and secure data handling.

Source Code Forensics & Software Forensics

Source code forensics and software forensics for litigation, internal investigations, and IP disputes. Code authorship attribution, commit history reconstruction, intellectual property lineage, and source code audit across repositories of any size.

Source code audit and forensics engagements include:

  • Code authorship attribution using commit-level analysis and git blame reconstruction
  • Copyright header scanning and license detection across entire source trees
  • Similarity and clone detection between competing codebases — relevant for trade secret and copyright cases
  • Reconstruction of code evolution and contributor history over multi-year timeframes
  • Invoice-to-deliverable reconciliation for contract and payment disputes
  • Repository-wide search and pattern detection for specific functions, identifiers, or text strings

Scale: Prior source code forensics engagements have covered hundreds of Git repositories, millions of commits, and hundreds of thousands of individual source files. Repository size is not a limiting factor — source code audit scope scales with the work, not the infrastructure.

Software, Firmware & Kernel Engineering (AI-Directed)

Software development, firmware engineering, and kernel-driver work — for deep technical projects that demand modern AI-assisted tooling to be feasible at solo scale. Peter directs Claude Code Max under careful engineering judgment to deliver scope that used to require a small team, at solo-practice pricing.

Engagements include:

  • Embedded firmware (NASM, C, Rust, ESP-IDF) for custom hardware
  • Device drivers and bare-metal code for non-standard targets
  • Kernel-level debugging, instrumentation, and hardware bring-up
  • Low-level systems code where compiler choices and memory layout matter
  • Large-codebase refactoring and legacy modernization via AI-directed rapid iteration
  • Custom build systems, CI/CD pipelines, and automated test harnesses
  • Automation tools, data pipelines, and integration work

How it works: Peter defines the architecture, picks the tools, and sets the test criteria. Claude Code Max writes and iterates on the code under that judgment. Every output is verified against real hardware or real data before it ships. Engineering decisions — what to build, what to test, what to release — remain Peter's. The AI handles implementation at a velocity that makes solo firmware engagements practically feasible. See the PaOS research project in Proven Results below for a concrete example of this workflow applied at 100K-line scale.

How It Works

Simple, transparent process from first call to final deliverable.

1

Tell Us What You Need

Describe your project, data set, or analysis requirements. Free 30-minute consultation to scope the work and give you a clear estimate.

2

We Analyze Your Data

Combining deep technical expertise with modern analytical tools, we process your documents, source code, or data thoroughly and efficiently.

3

Get Verified Results

Every deliverable passes through automated quality verification before it ships. Receive clear, well-organized work product ready for legal proceedings, board presentations, or internal decision-making.

Proven Results

Real engagements and research projects demonstrating what thorough technical consulting can deliver.

Case Study

IP Ownership Dispute — Large-Scale Corporate Archive to Court-Filed Expert Report

An investment firm needed to establish that their portfolio company — not the customer — owned the intellectual property in a jointly-developed product. The data set included terabytes of corporate archives spanning multiple years. The work evolved from data analysis into a court-filed expert report for a federal IP case.

TB-scale Data analyzed
300+ Git repos reviewed
70+ Reports produced
<100 hrs Total time

What we did:

  • Built a semantic search index across hundreds of thousands of files for full-archive analysis
  • Performed source code forensics across all repositories — millions of commits analyzed for authorship attribution
  • Scanned hundreds of thousands of source files for copyright headers and reconstructed multi-year contract history
  • Reconciled dozens of invoices totaling millions in contract payments
  • Prepared a comprehensive expert report with supporting exhibits, filed with the court under Daubert standards
  • Every deliverable verified by automated quality assurance before shipping

Expert report accepted by the court. This scope of work would typically require a team of analysts working several months.

Case Study

CNC Machine Controller Redesign — Ground-Up Embedded System for Production Manufacturing

A CNC equipment manufacturer needed a complete redesign of their machine controller. The existing design — built around two processors and 29 optoisolators — was expensive to manufacture, difficult to maintain, and couldn’t support the wireless control or expandability the company wanted. Working closely with the client’s lead engineer, the engagement spanned seven years: from initial research and prototyping through multiple PCB revisions, firmware development, and production documentation for volumes of 5,000–10,000 units per year.

7 years Engagement duration
6+ Distinct PCB designs
5 BLDC motor axes
10+ Accessory modules designed

What we did:

  • Replaced the legacy dual-processor architecture with an ESP32-based design (dual-core 240MHz) plus an auxiliary 8-bit microcontroller for the accessory subsystem
  • Evaluated and selected motor driver ICs for 5-axis brushless DC motor control at 48V, including thermal analysis, load testing, and discrete FET evaluation
  • Designed and tested the power delivery system with regulated 48V, 24V, 5V, and 3.3V rails, including PFC research and extensive bench testing with adjustable loads and precision supplies
  • Built the plasma torch interface: 240VAC torch fire control, arc voltage sensing, touch-off detection, and integration with commercial plasma cutting systems
  • Developed embedded firmware (ESP-IDF) with over-the-air update capability, WiFi host communications, and a wireless control interface — after extensive Bluetooth prototyping and latency testing proved WiFi was the better path for Windows compatibility
  • Designed a modular accessory ecosystem with marking laser integration, each module using its own I2C-connected controller board
  • Created full PCB designs (schematic, layout, Gerber files, BOM) across 6+ boards including the main controller, torch interface, evaluation boards, and accessory kit
  • Produced factory test procedures (3-stage: PCB self-check, load test/calibration, full table integration), failure mode analysis, and production manufacturing documentation

Helped deliver a production-ready controller platform that replaced the legacy hardware with a modern, wireless-capable design supporting a full accessory ecosystem — work spanning component research, schematic capture, PCB layout, firmware, power supply testing, and manufacturing documentation.

Research Project

PaOS — Bare-Metal Operating System via AI-Directed Firmware Development

PaOS is an ongoing proprietary research project: a 64-bit UEFI bare-metal operating system written in pure NASM assembly, developed through a continuous AI-directed workflow. It's the proof point for what solo firmware engineering looks like when you know how to orchestrate modern AI development tools on a real problem at real scale.

100K+ Lines of NASM assembly
370+ Tagged releases
3 Bare-metal hardware targets

What the project demonstrates:

  • PE32+ UEFI application with architected boot sequence from cold power-on through full desktop environment — no BIOS, no GRUB, no C toolchain
  • Hardware drivers developed iteratively and verified on real metal: XHCI USB 3.0 stack, Ethernet, NVMe storage, framebuffer graphics, interrupt handling, UEFI protocols, and serial I/O
  • On-device LLM inference running inside the kernel address space
  • Continuous bare-metal deploy loop: automated builds, SSH-based firmware updates to Beelink EQi12, regression capture via UDP, per-commit hardware verification
  • 546-task roadmap with every task gated on bare-metal test passes — not just QEMU simulation
  • Automated regression harness with hundreds of serial-output markers across graphics, networking, storage, interrupts, and UEFI protocol behavior
  • Fault-tolerant design: graceful hardware degradation, task isolation with watchdog, kernel panic recovery with user-facing reboot prompts

The workflow: Peter defines architecture, target behavior, and test criteria. Claude Code Max writes the assembly and runs the QEMU regression suite. A deploy pipeline pushes new builds to three bare-metal targets automatically and captures serial output to verify the code actually boots on real hardware (QEMU is necessary but not sufficient). Peter reviews outputs, iterates on anything that fails bare-metal verification, and tags a release only when all targets pass. Engineering judgment stays with Peter; the AI handles implementation at a velocity that wouldn't be feasible otherwise.

100K+ lines of working assembly across 370+ tagged releases, running on three different bare-metal targets. A single developer shipping production-quality firmware at a velocity that used to require a small embedded systems team — because the AI-directed workflow makes it feasible. The same workflow translates directly to client firmware, kernel-driver, and deep systems engineering projects.

About

Deep technical expertise and rigorous methodology, delivering thorough results efficiently.

Gransee Engineering is a solo technical consulting practice founded by Peter Gransee and based in Phoenix, Arizona. We serve clients nationwide.

Peter brings 25 years of hands-on technical experience across software engineering, Linux systems, data analysis, infrastructure, and AI/ML. He's built and run 6 businesses — including a flashlight company that grew to $2M+ in sales and 11 employees — and has worked with everything from embedded firmware to enterprise cloud platforms.

Modern analytical tools — including custom search indexes, automated quality checks, and scripted analysis pipelines — enhance the work, but they don't replace expertise. Every conclusion is reviewed and validated by Peter personally. That means faster turnaround, lower costs, and direct access to the person actually doing the work.

Every engagement includes a confidentiality agreement. Your data stays secure, and you always know exactly how AI tools are being used in your project.

25+ Years of technical experience
6 Businesses founded
$2M+ In sales — first business
Fast Turnaround with modern tools

Transparent Pricing

No minimums, no hidden fees, no per-GB charges. You know exactly what you're paying for.

Service Hourly Rate
Document review & analysis $200/hr
Software engineering & automation $200/hr
Source code forensics $225/hr
Expert report preparation $225/hr
Technical due diligence $250/hr
Expert testimony / deposition $350/hr

Let's Talk

Describe what you need and we'll get back to you within one business day. Initial 30-minute consultations are always free.

Email
peter@gransee.com
Location
Phoenix, AZ — serving clients nationwide