---

name: juicebox-migration-deep-dive description: | Advanced Juicebox data migration strategies. Use when migrating from other recruiting platforms, performing bulk data imports, or implementing complex data transformation pipelines. Trigger with phrases like "juicebox data migration", "migrate to juicebox", "juicebox import", "juicebox bulk migration". allowed-tools: Read, Write, Edit, Bash(kubectl:), Bash(curl:) version: 1.0.0 license: MIT author: Jeremy Longshore jeremy@intentsolutions.io

Juicebox Migration Deep Dive

Overview

Advanced strategies for migrating data to Juicebox from other recruiting and people search platforms.

Prerequisites

• Source data access and export capabilities
• Juicebox Enterprise plan (for bulk imports)
• Data mapping documentation
• Testing environment

Migration Sources

Instructions

Step 1: Data Assessment

Copy// scripts/assess-source-data.ts
interface DataAssessment {
  totalRecords: number;
  uniqueProfiles: number;
  duplicates: number;
  fieldCoverage: Record<string, number>;
  dataQualityScore: number;
  estimatedMigrationTime: string;
}

export async function assessSourceData(
  source: string,
  sampleSize: number = 1000
): Promise<DataAssessment> {
  const sample = await loadSampleData(source, sampleSize);

  const assessment: DataAssessment = {
    totalRecords: sample.total,
    uniqueProfiles: new Set(sample.records.map(r => r.email)).size,
    duplicates: sample.total - new Set(sample.records.map(r => r.email)).size,
    fieldCoverage: calculateFieldCoverage(sample.records),
    dataQualityScore: calculateQualityScore(sample.records),
    estimatedMigrationTime: estimateMigrationTime(sample.total)
  };

  return assessment;
}

function calculateFieldCoverage(records: any[]): Record<string, number> {
  const fields = ['name', 'email', 'title', 'company', 'location', 'phone'];
  const coverage: Record<string, number> = {};

  for (const field of fields) {
    const count = records.filter(r => r[field] && r[field].trim()).length;
    coverage[field] = (count / records.length) * 100;
  }

  return coverage;
}

Step 2: Schema Mapping

Copy// lib/migration/schema-mapper.ts
export interface FieldMapping {
  sourceField: string;
  targetField: string;
  transform?: (value: any) => any;
  required: boolean;
}

export const linkedInMapping: FieldMapping[] = [
  { sourceField: 'firstName', targetField: 'first_name', required: true },
  { sourceField: 'lastName', targetField: 'last_name', required: true },
  {
    sourceField: 'fullName',
    targetField: 'name',
    transform: (v) => v || undefined,
    required: false
  },
  { sourceField: 'headline', targetField: 'title', required: false },
  { sourceField: 'companyName', targetField: 'company', required: false },
  {
    sourceField: 'location',
    targetField: 'location',
    transform: normalizeLocation,
    required: false
  },
  {
    sourceField: 'profileUrl',
    targetField: 'linkedin_url',
    transform: normalizeLinkedInUrl,
    required: false
  },
  {
    sourceField: 'connectionDegree',
    targetField: 'metadata.connection_degree',
    required: false
  }
];

export class SchemaMapper {
  constructor(private mappings: FieldMapping[]) {}

  mapRecord(source: Record<string, any>): Record<string, any> {
    const target: Record<string, any> = {};

    for (const mapping of this.mappings) {
      let value = this.getNestedValue(source, mapping.sourceField);

      if (mapping.transform) {
        value = mapping.transform(value);
      }

      if (value !== undefined && value !== null && value !== '') {
        this.setNestedValue(target, mapping.targetField, value);
      } else if (mapping.required) {
        throw new Error(`Required field missing: ${mapping.sourceField}`);
      }
    }

    return target;
  }
}

Step 3: Data Transformation Pipeline

Copy// lib/migration/pipeline.ts
import { Transform, pipeline } from 'stream';
import { promisify } from 'util';

const pipelineAsync = promisify(pipeline);

export class MigrationPipeline {
  private stages: Transform[] = [];

  addStage(name: string, transform: (record: any) => any): this {
    this.stages.push(new Transform({
      objectMode: true,
      transform(record, encoding, callback) {
        try {
          const result = transform(record);
          if (result) {
            this.push(result);
          }
          callback();
        } catch (error) {
          callback(error as Error);
        }
      }
    }));
    return this;
  }

  async run(source: Readable, destination: Writable): Promise<MigrationStats> {
    const stats = new MigrationStats();

    const statsTracker = new Transform({
      objectMode: true,
      transform(record, encoding, callback) {
        stats.increment();
        this.push(record);
        callback();
      }
    });

    await pipelineAsync(
      source,
      ...this.stages,
      statsTracker,
      destination
    );

    return stats;
  }
}

// Usage
const pipeline = new MigrationPipeline()
  .addStage('parse', parseCSVRecord)
  .addStage('validate', validateRecord)
  .addStage('deduplicate', deduplicateRecord)
  .addStage('transform', transformToJuiceboxSchema)
  .addStage('enrich', enrichWithMetadata);

Step 4: Bulk Import with Rate Limiting

Copy// lib/migration/bulk-importer.ts
export class BulkImporter {
  private rateLimiter: RateLimiter;
  private batchSize: number;
  private maxConcurrent: number;

  constructor(options: {
    requestsPerSecond: number;
    batchSize: number;
    maxConcurrent: number;
  }) {
    this.rateLimiter = new RateLimiter(options.requestsPerSecond);
    this.batchSize = options.batchSize;
    this.maxConcurrent = options.maxConcurrent;
  }

  async import(records: Profile[]): Promise<ImportResult> {
    const result: ImportResult = {
      total: records.length,
      successful: 0,
      failed: 0,
      errors: []
    };

    // Split into batches
    const batches = chunk(records, this.batchSize);

    // Process batches with concurrency limit
    const semaphore = new Semaphore(this.maxConcurrent);

    await Promise.all(batches.map(async (batch, index) => {
      await semaphore.acquire();
      try {
        await this.rateLimiter.wait();
        const batchResult = await this.importBatch(batch);

        result.successful += batchResult.successful;
        result.failed += batchResult.failed;
        result.errors.push(...batchResult.errors);

        logger.info(`Batch ${index + 1}/${batches.length} complete`, {
          successful: batchResult.successful,
          failed: batchResult.failed
        });
      } finally {
        semaphore.release();
      }
    }));

    return result;
  }

  private async importBatch(batch: Profile[]): Promise<BatchResult> {
    try {
      const response = await juiceboxClient.profiles.bulkImport(batch);
      return {
        successful: response.created + response.updated,
        failed: response.failed,
        errors: response.errors
      };
    } catch (error) {
      return {
        successful: 0,
        failed: batch.length,
        errors: [{ message: (error as Error).message, records: batch }]
      };
    }
  }
}

Step 5: Validation and Reconciliation

Copy// lib/migration/validator.ts
export class MigrationValidator {
  async validateMigration(
    sourceCount: number,
    destinationQuery: string
  ): Promise<ValidationReport> {
    const report: ValidationReport = {
      sourceCount,
      destinationCount: 0,
      matchRate: 0,
      missingRecords: [],
      dataIntegrityIssues: []
    };

    // Count destination records
    const destResult = await juiceboxClient.search.people({
      query: destinationQuery,
      limit: 0
    });
    report.destinationCount = destResult.total;
    report.matchRate = (report.destinationCount / sourceCount) * 100;

    // Sample validation
    const sampleSize = Math.min(100, sourceCount);
    const sample = await this.getSampleFromSource(sampleSize);

    for (const record of sample) {
      const match = await this.findInDestination(record);
      if (!match) {
        report.missingRecords.push(record.id);
      } else {
        const issues = this.compareRecords(record, match);
        if (issues.length > 0) {
          report.dataIntegrityIssues.push({
            recordId: record.id,
            issues
          });
        }
      }
    }

    return report;
  }

  private compareRecords(source: any, dest: any): string[] {
    const issues: string[] = [];
    const criticalFields = ['name', 'email', 'company'];

    for (const field of criticalFields) {
      if (source[field] !== dest[field]) {
        issues.push(`${field} mismatch: "${source[field]}" vs "${dest[field]}"`);
      }
    }

    return issues;
  }
}

Step 6: Rollback Strategy

Copy// lib/migration/rollback.ts
export class MigrationRollback {
  private checkpointFile: string;

  constructor(migrationId: string) {
    this.checkpointFile = `./checkpoints/${migrationId}.json`;
  }

  async saveCheckpoint(state: MigrationState): Promise<void> {
    await fs.writeFile(this.checkpointFile, JSON.stringify(state, null, 2));
  }

  async loadCheckpoint(): Promise<MigrationState | null> {
    try {
      const data = await fs.readFile(this.checkpointFile, 'utf-8');
      return JSON.parse(data);
    } catch {
      return null;
    }
  }

  async rollback(migrationId: string): Promise<RollbackResult> {
    const checkpoint = await this.loadCheckpoint();
    if (!checkpoint) {
      throw new Error('No checkpoint found for rollback');
    }

    // Delete imported records
    const deleted = await juiceboxClient.profiles.bulkDelete({
      filter: { migrationId }
    });

    return {
      recordsRolledBack: deleted.count,
      checkpoint: checkpoint.lastProcessedId
    };
  }
}

Migration Checklist

Copy## Pre-Migration
- [ ] Source data exported and validated
- [ ] Field mapping documented
- [ ] Test migration on sample data
- [ ] Rollback plan documented
- [ ] Stakeholder sign-off

## During Migration
- [ ] Monitoring dashboards active
- [ ] Progress tracking enabled
- [ ] Error logging configured
- [ ] Checkpoint saves working

## Post-Migration
- [ ] Reconciliation complete
- [ ] Data integrity verified
- [ ] Source system archived
- [ ] Documentation updated
- [ ] Team training complete

Output

• Data assessment tools
• Schema mapping configuration
• Transformation pipeline
• Bulk import with rate limiting
• Validation and reconciliation

Resources

• Bulk Import Guide
• Data Format Specifications

Summary

This skill pack completes the enterprise-grade Juicebox integration toolkit.