Back to Dashboard
Jordan Kim
Full Stack Developer · LegalZoom
Borderline
Screening
Completed 2026-02-20
Agentic Score
58
/ 100
Time
89 min
vs 30 min par
Tool Calls
8
API invocations
Plan Steps
2
decomposition depth
Cohort Rank
#4
of 5 candidates
Key Highlight
Struggled to decompose the eDiscovery task into sub-problems
Evaluator Notes
Good coding instincts but reactive rather than proactive. No upfront planning.
Score Breakdown
Planning
45Tool Use
72Reasoning
58Logical AnalysisDomain Knowledge
Adaptation
49ReflectionIteration Speed
Communication
68Agentic Radar
vs. Industry Median
Planning
61-16
Tool Use
58+14
Reasoning
63-5
Adaptation
55-6
Communication
67+1