By Michael Barr
Note: This article offers practical guidance for attorneys engaging technical experts. It is general commentary, not legal advice. Consult counsel about any specific matter.
I once spent the better part of two days on the witness stand in an Oklahoma courtroom, explaining to a jury what my team had found inside the engine control software of a Toyota Camry. The jurors were not engineers. That was the point. My job was to make concepts like task death and stack overflow feel as concrete as a stuck throttle, because in district court the technical expert's first job is to teach.
Move the same dispute to the Patent Trial and Appeal Board and that job disappears. The judges there are technically trained, many in the very field of the patent. They do not want the tutorial. They will grade the expert on rigor, and nothing else. Move it to the International Trade Commission and the job changes a third time: more to prove, in less time, for an audience that lives in patent cases.
Litigators know these forums differ procedurally. What still surprises many is how differently the three forums use the technical expert, and how often an expert who is superb in one is merely adequate in another.
The short version: district court needs a teacher who can survive Daubert in front of a lay jury. The PTAB needs a prior-art specialist who can defend an obviousness case to technical judges inside a statutory one-year clock. The ITC needs breadth and stamina: infringement, validity, and the domestic industry technical prong, all at once, on the fastest schedule in patent practice. Choose the expert for the forum, not for the resume. And keep one eye on the quiet fourth track, ex parte reexamination at the Patent Office, which can pause the whole fight.
District court: teaching a jury, surviving Daubert
District court is the forum most people picture. A federal judge, usually a jury, a full trial, a schedule measured in years.
The expert here carries the heaviest and broadest load: infringement or non-infringement, often validity, support for claim construction at the Markman stage. In a software or electronics case, a source code review sits underneath most of it. The expert, directing a review team, maps the accused code to the claim limitations, writes a Rule 26 report, sits for a long deposition, and then has to explain all of it to twelve people who have never written a line of code.
Here is what the witness stand teaches you that no resume shows. A jury decides whether they follow you and whether they trust you, in that order, and they decide the second one fast. The most technically brilliant expert in the field is the wrong choice if that expert cannot make memory management feel intuitive without saying anything a cross-examiner can catch as false. Depth plus the ability to teach is a rarer combination than most people assume, and you cannot verify it from a CV. You have to hear the candidate explain something hard.
Before the jury hears a word, there is Daubert. The opinions must survive a challenge to their reliability and methodology, and what carries them through that gate is the discipline underneath: a defensible, documented code review and a method the expert can articulate under oath. This is the same discipline we cover in our guide to vetting a source code review team. The review beneath the opinion determines whether the opinion holds.
One more number matters for what comes next: invalidity in district court must be proven by clear and convincing evidence. Remember that standard. The PTAB does not use it.
The PTAB: proving invalidity to engineers, fast
The Patent Trial and Appeal Board, inside the U.S. Patent and Trademark Office, hears validity challenges, overwhelmingly through inter partes review (IPR). (Post-grant review exists for patents less than nine months old, but IPR is the workhorse, so IPR is what this section describes.) No jury. No district judge. A panel of administrative patent judges, technically trained, frequently in the field of the patent.
The work is different in kind, not just in audience. An IPR attacks the patent as anticipated or obvious over prior art under sections 102 and 103, so the expert's craft is prior-art analysis: reconstructing what a person of ordinary skill knew years ago and defending an obviousness combination against attack. I have seen excellent code-review experts struggle here, because reading accused source code and reconstructing the state of the art in 2011 are different skills that happen to live under one word, "expert."
The burden drops to preponderance of the evidence, a large part of why accused infringers so often file an IPR in parallel: a second, easier shot at the same patent, in front of a more technical audience. And the clock is statutory. The Board issues its final written decision within one year of institution, extendable six months for good cause. Declarations are front-loaded, depositions land inside the window, and an expert without room in the calendar is an expert who will miss it.
A trap worth naming plainly: claim construction at the PTAB now follows the same Phillips standard as district court, so a construction taken in one forum travels to the other. An expert who takes inconsistent positions across parallel proceedings has handed opposing counsel the best exhibit they will ever get.
The ITC: domestic industry, and a brutal schedule
A Section 337 investigation at the U.S. International Trade Commission is built for one thing: stopping infringing imports at the border. The remedy is not money. It is an exclusion order, plus cease-and-desist orders. An Administrative Law Judge hears the case, there is no jury, and the target date is commonly sixteen to eighteen months from institution, with discovery at a pace many litigators describe as the most punishing in patent practice.
The ITC also adds a technical demand the other forums do not have: domestic industry. To get relief, a complainant must show that its own products actually practice the asserted patent (the technical prong; the economic prong usually calls for a separate economics expert). So one technical expert may need to prove three things at once: the imports infringe, the patent is valid, and the complainant's own products practice it. That is a wider mandate than either district court or the PTAB, on the tightest schedule of the three.
Two selection consequences follow, and neither is negotiable. The expert and the code review team behind the expert must be able to commit intensively; the ITC schedule does not bend around anyone's other matters, and an expert stretched across several cases is a real risk here in a way a district court schedule can absorb. And the ALJ is patent-sophisticated, so rigor and command of the record beat showmanship, while clarity still counts because the record is built for review.
The quiet fourth track: ex parte reexamination
There is a fourth way to attack a patent, and we are seeing it more often. Ex parte reexamination is not a forum in the sense the other three are. There is no trial, no deposition, no cross-examination. Anyone, including an anonymous party or the patent owner itself, can ask the Patent Office at any time during the life of a patent to take a second look based on patents and printed publications the examiner did not fully consider. The threshold is a "substantial new question of patentability," and the Office grants the overwhelming majority of requests. After the request is filed, the challenger steps out of the room: the proceeding runs strictly between the patent owner and a Patent Office examiner in the Central Reexamination Unit.
Three properties make it strategically interesting. It costs a small fraction of an IPR. It can be filed anonymously. And it carries no statutory estoppel, so a defendant who loses at the reexamination stage is not barred from raising the same art in court later. Patent owners use it too, going in voluntarily to amend and tighten claims before a fight they see coming.
Do not mistake quiet for harmless. In May 2026, in Municipal Parking Services v. Parking Revenue Recovery Services in the Western District of Texas, Judge Albright granted an unopposed motion and stayed the entire case pending completion of ex parte reexaminations. That is the busiest patent venue in the country pausing a case that had already been through its Markman hearing, on the patent owner's own motion. When a reexamination can freeze a district court case after claim construction, it belongs in your planning.
The expert's role here is narrow but real. A reexamination request lives or dies on the quality of its prior-art story, and a supporting expert declaration explaining what a person of ordinary skill would have understood from the references is often what makes that story credible. On the patent owner's side, declarations supporting amended claims call for the same prior-art fluency the PTAB demands, delivered in writing to an examiner rather than under cross-examination. The expert who fits this work is the PTAB-style scholar, not the jury-trial teacher, and the writing has to stand entirely on its own, because nobody will be there to defend it orally.
Seeing them side by side
| District Court | PTAB (IPR) | ITC (Section 337) | Ex Parte Reexam | |
|---|---|---|---|---|
| Decision-maker | Judge and usually a jury | Panel of technical patent judges | Administrative Law Judge | Patent Office examiner (Central Reexamination Unit) |
| What the expert proves | Infringement, often validity, claim construction | Invalidity over prior art (obviousness, anticipation) | Infringement, validity, and domestic industry technical prong | Prior-art case in a written declaration, or support for amended claims |
| Invalidity burden | Clear and convincing evidence | Preponderance of the evidence | Clear and convincing evidence | Substantial new question to open; then examination, not litigation |
| Remedy at stake | Damages, injunction | Patent claims canceled or upheld | Exclusion order, no damages | Claims canceled, amended, or confirmed |
| Typical timeline | Years | About 12 to 18 months | About 16 to 18 months, fastest discovery | Typically about two years to a certificate |
| Audience for the expert | Lay jury, teach from the ground up | Technical judges, show rigor | Patent-savvy ALJ, rigor plus clarity | An examiner, in writing only; nobody to defend it orally |
The most important row is not in the table: these proceedings run in parallel on the same patents all the time. The accused infringer sued in district court files an IPR. The ITC respondent is litigating the same patent in district court, which may stay its case pending the ITC result. When that happens, every transcript and declaration gets read by opposing counsel hunting for one thing: a contradiction. One expert, or one tightly coordinated team, carrying consistent positions across all forums is usually the safest way to keep the technical story straight.
A checklist for matching the expert to the forum
-
Name the forum, then the job. Teaching a jury, defending an obviousness case, or proving domestic industry at a sprint are different jobs. Start there, not with the CV.
-
For a jury forum, test teaching ability directly. Ask the candidate to explain a hard piece of the accused technology the way they would to a jury. If your paralegal cannot follow it, neither will twelve jurors.
-
For the PTAB, probe prior-art and obviousness experience specifically: declarations written, combinations defended, Board depositions survived. Code-reading brilliance does not transfer automatically.
-
For the ITC, confirm bandwidth first and domestic-industry experience second. The schedule is unforgiving and an overcommitted expert will fail you at the worst moment.
-
In every forum, check the code review underneath. Infringement and domestic industry both rest on a defensible, well-documented review. The audience changes; the discipline does not. See our guide to vetting a code review team.
-
Plan for parallel proceedings from the start. If the same patent is in, or may end up in, more than one forum, favor an expert or coordinated team who understands how a construction taken in one forum travels to the others.
-
Ask about the reexamination angle, on both sides. If a quiet ex parte reexamination could challenge the patent, or strengthen it before trial, the expert plan needs a prior-art scholar whose written declaration can stand on its own, and a schedule that survives a stay.
What we do at Barr Group
Our electronics and software expert witnesses work across all three forums, and we match the expert to the forum and the technology rather than sending the same person everywhere by default. For a jury trial, that means experts who combine real engineering depth with the ability to teach and hold up under Daubert. For the PTAB, experts fluent in prior-art analysis who write declarations for technical judges on a compressed schedule. For the ITC, experts and source code review teams who can absorb the pace and cover infringement, validity, and the domestic industry prong together. And when one patent is being fought on multiple fronts, we keep the technical opinions consistent across all of them, so nothing can be turned against the client.
The forum sets the job
There is no single "patent expert" who is automatically right for every proceeding. A jury trial rewards a teacher. A PTAB review rewards a scholar on a deadline. An ITC investigation rewards breadth and stamina. I have watched all three up close, and the pattern holds: the experts who fail are rarely underqualified, they are mismatched. Choose for the forum you are actually in, plan early for the ones you might add, and insist on a documented code review beneath all of it.
Barr Group's team of electronics and software expert witnesses provide experienced and unbiased source code reviews, expert reports and testimony for product liability, patent infringement, software copyright, and trade secrets litigation involving computer-based technology and software. HIRE AN EXPERT
Get a matched recommendation >
Free, confidential, conflicts-checked, and our fee is built into the expert's hourly rate, so we are paid only if you retain.