Fintech / Enterprise Backend — 2026

Revela Africa

Enterprise vehicle liquidation & AI valuation backend with asynchronous BullMQ job queues.

revela-africa.typescript

Core Implementation Snippet

// Revela Africa BullMQ AI Valuation Processor
@Processor('valuation-queue')
export class ValuationProcessor extends WorkerHost {
  @Process('compute-valuation')
  async handleValuation(job: Job<{ vehicleId: string; specs: VehicleSpecs }>) {
    const { vehicleId, specs } = job.data;
    const valuation = await this.aiService.estimateFairMarketValue(specs);
    await this.vehicleRepo.update(vehicleId, {
      estimatedValue: valuation.amount,
      status: ValuationStatus.PENDING_INSPECTION
    });
    return valuation;
  }
}

Queue EngineBullMQ + Redis

API ProtocolGraphQL Code-First

SecurityJWT & Nest Guards

DatabasePostgreSQL

Role

Core Backend Engineer

Timeline

2026

Technologies

NestJSGraphQLPostgreSQLBullMQRedisTypeScriptDockerJWT

Overview & Problem Solved

The core backend powering a multi-stage vehicle liquidation ecosystem, orchestrating the entire lifecycle from AI-assisted valuation to on-site inspector verification, admin approvals, and merchant settlements.

System Architecture

Architected in NestJS, GraphQL (code-first), PostgreSQL, and BullMQ with Redis for background valuation job distribution. Secured with JWT, NestJS Guards, and role-based access control (RBAC).

Key Outcomes

▪Decoupled heavy AI computation from client request cycles using BullMQ + Redis job queues with automatic retry logic.
▪Designed multi-stage inspection state machine: AI Valuation → Inspector Review → Admin Final Offer.
▪Eliminated overfetching across vendor dashboards via code-first GraphQL resolvers.

← Previous ProjectSlaviyy Next Project →Asoebi Africa