Legal Expert

About this role

# Domain Expert – Legal (PACER / Federal Court Records) To qualify you must use **PACER (and federal court dockets / CM/ECF)** regularly, weekly or more, as a working part of your job, with **2+ years of professional federal-litigation experience** as a litigation paralegal, docket/court clerk, legal-research librarian, or litigator. ## **PACER knowledge we require** **Reading cases & dockets** - Reading a docket sheet fluently: parties, attorneys, assigned judge, filed/terminated dates, case status - Docket entries: interpreting terse entry text, following cross-referenced entries, and tracing procedural posture (what was filed, how the judge ruled, how the case ended) - Decoding case numbers (division, year, case type, judge initials) and recognizing what they tell you **Searching & finding** - PACER Case Locator (nationwide index) vs. searching inside a specific court's CM/ECF, and the limits of each - Party-name search and its traps: name variants, corporate suffixes, "et al.", d/b/a, subsidiaries, misspellings - Filtering by court, date range, case type, and Nature of Suit; targeted retrieval instead of blind browsing **Court structure & jurisdiction** - The federal court landscape: district, appellate (circuit), and bankruptcy courts, and knowing which court a case lives in - Case relationships: consolidated cases, MDL lead vs. member cases, transfers, and district-to-appeal linkage **Case types & classification** - Civil / criminal / bankruptcy case types; Nature of Suit codes; cause of action; party roles (plaintiff, defendant, appellant/appellee, debtor/creditor, trustee) **Documents, records & workflow** - Document types: complaint, answer, motion, brief, order, judgment, exhibit, and what each represents - What is and is not retrievable: sealed cases, restricted/redacted documents, minute entries with no document, and the privacy rules behind them - App workflow surfaces: saved searches, saved cases, and batch download jobs for large result sets ## **What you'll do** - **Review and QA AI-agent rollouts:** work through completed agent runs and judge whether the agent's answer is correct against the actual docket, right case, right court, right parties, right outcome/status, right document - **Identify failure modes:** pinpoint where and why an agent succeeds too easily or fails wrongly, and flag tasks that are too easy or solvable without actually using PACER - **Improve realism and difficulty:** refine the task prompt and the seeded world so tasks reflect authentic federal docket research and are genuinely hard for a capable agent - **Update grading guidance for accuracy:** sharpen how answers are graded so correct answers pass and wrong ones fail, including gating failures (wrong case identified, misread procedural posture, missed member case, or use of a sealed/restricted document that should be off-limits) - **Catch plausible-but-wrong answers** that a non-expert reviewer would let slide ## **Requirements** - Frequent hands-on PACER / CM/ECF use for federal docket research in your current or recent role, weekly or more - 2+ years in federal litigation support: litigation paralegal, docket/court clerk, legal-research librarian, or litigator - Can read a federal docket on sight: decode a case number, trace procedural posture, tell open vs. terminated, recognize case types and Nature of Suit - Reflexive with PACER search mechanics and the name-variant traps that cause missed cases - Sound judgment on what a correct, complete answer looks like, and the confidence to mark a plausible answer wrong - Bonus: bankruptcy docket experience; complex/multidistrict litigation; hands-on experience across multiple courts (not just one firm's home district) - Note: a J.D. and bar admission are welcome but **not required**; experienced litigation paralegals and docket specialists are the primary pool

Skills & domains

• ai-training
• rlhf
• sme
• annotation
• Law