How to grouping and aggregation using Gremlin

Question

1. we have part vertex and job vertex
2. parts vertex linked to job vertex with relationship 'parts used in job'
3. I have a set of parts as input say 10 parts. For these 10 parts can you identify possible jobs. The output I need is like job1: "[Part1,Part2,part3]" job2: "[part2,part4]"

{"gremlin":"g.V().hasLabel('Parts').has('~id', within('2197001','6D9157','6V8636','336027','6V3831','2554534','2531351')).as('part').out('part_used_in_job').as('job').select('job', 'part').group().by('job').by(select('part').values('~id').fold())"} {'detailedMessage': 'java.util.LinkedHashMap cannot be cast to org.apache.tinkerpop.gremlin.structure.Element', 'requestId': '49551d9e-e22f-4e92-99bc-77fbf1b1784b', 'code': 'UnsupportedOperationException', 'message': 'java.util.LinkedHashMap cannot be cast to org.apache.tinkerpop.gremlin.structure.Element'} """

g.V().hasLabel('Parts').has('~id', within('2197001','6D9157','6V8636','336027','6V3831','2554534','2531351'))
  .as('part')
  .out('part_used_in_job')
  .as('job')
  .select('job', 'part')
  .group()
    .by('job')
    .by(select('part').values('~id').fold())

"""