FDE Hub logo: bright terminal prompt and forward arrow joined at a hub nodeFDEHUB.DEV

Interview

Harvey AI FDE interview guide

Harvey Forward Deployed loops test whether you can land legal AI workflows under confidentiality, accuracy, and firm operating constraints — trust and citation discipline matter as much as model fluency.

Legal vertical seats punish casual hallucination tolerance. Interviewers listen for whether you design review gates and measurable quality before promising leverage.

Interview loop matrix

StageWhat they probeFormatPass signal
Recruiter screenDomain humility, coding bar, travel, client exposure30–45 min callYou frame delivery ownership without claiming to be counsel
Technical screenSystems, RAG, APIs, debuggingExercise or discussionGrounded design under incomplete corpora
Legal workflow caseScoping, matter types, review gates, stakeholdersAmbiguous scenarioOne practice-group thin slice with human review
Trust / confidentiality deep diveData boundaries, citations, auditabilityDesign + critiqueConfidentiality and provenance as design inputs
Behavioral / HMPartner communication, conflict, ambiguityStory-drivenPrecise risk language; no overclaiming advice

Grading rubric

DimensionStrongWeak
Domain humilityScopes engineerable workflows; defers legal judgmentTalks like unauthorized counsel
ConfidentialityMatter isolation, access controls, audit trails by defaultShared corpora across sensitive boundaries
Grounding disciplineCitations, source checks, explicit uncertaintyFluent answers without provenance
Review gatesHuman review on high-stakes outputs before wider useAutonomous drafting with no escalation path
Scope controlOne matter type / one practice group / one KPIFirm-wide AI assistant day one
CommunicationPartner-ready risk language; clear promotion criteriaHides failure modes to protect the demo

Red flags

  • No human review path for high-stakes legal outputs
  • Ignoring matter confidentiality / data boundaries
  • Equating fluent drafting with verified accuracy
  • Cannot define a measurable pilot for one practice group

Practice scenarios

  1. Pilot for contract review assistants in one practice group — what is in vs out of scope for 30 days?
  2. An answer cites the wrong clause confidently — how do you diagnose retrieval vs generation and gate rollout?
  3. Partners want firm-wide enablement tomorrow — what evidence do you require first?

Pair with Enterprise RAG, agent evals, Glean FDE, and the Translation Matrix.

7-day prep plan

  1. Day 1–2 — Grounded RAG + citation discipline drill
  2. Day 3 — Confidentiality / matter isolation design
  3. Day 4 — Full practice-group pilot case
  4. Day 5 — Eval gates for high-stakes outputs
  5. Day 6 — Behavioral: partner communication under uncertainty
  6. Day 7 — Mock loop; score on the rubric

Comp context

Directional TC discussions often land around $190K–$380K. Details: Harvey AI FDE salary hub. Peers: Sierra, Glean.