Harvey
Software EngineerStrategicAug 6, 2026

Improve follow-up research continuation without misleading context

Iterative research often needs continuity without quietly turning earlier assumptions into constraints.

A useful thread can preserve momentum while an overly persistent one can narrow the next question too soon.

Sometimes I need the last answer in my head; sometimes I need it completely out.

Ingrid Bakker · Employment Law Partner

Builds evolving research threads while refining client advice across changing fact patterns.

What pulls against what

  • continuity vs. fresh reasoning
  • behavioral telemetry vs. lawyer intent
  • rapid learning vs. citation-quality safeguards
  • automatic context vs. user control

What is at stake

The visible drop-off has several plausible causes. The work is to find a reversible technical path that improves continuity without contaminating legal reasoning

Why Harvey

At Harvey, this can matter because research work often develops through successive questions rather than a single prompt.

Written for

Experiment-minded engineerSearch and retrieval engineerProduct systems engineer

This is the setup. The work is inside.

Running it puts you in the room: the full situation and its constraints, stakeholders who push back in their own words, and the decisions that are yours to make. What you produce becomes a Day One Plan — work you can show someone instead of describing.