Search...

AI Developer

VAAS logo
VAAS

VAAS provides an intelligent risk-management platform for compliance, fraud prevention, and credit decisions. It serves risk teams at fintechs, banks, insurers, industrial companies, and retailers.

Distributed
About VAAS

VAAS is a risk intelligence platform that helps organizations make secure decisions in seconds across compliance, fraud prevention, and credit. Its product includes configurable workflows, a decision desk, AI risk agents, continuous risk monitoring, and a data hub that integrates risk data sources. The platform supports onboarding and monitoring of people and businesses, real-time fraud detection, and explainable credit policies for teams in fintech, banking, insurance, industry, and retail.

View jobs by VAAS

Skills

Candidate Availability

Required and preferred rules are kept separate and reflect the wording in the original posting.

About the Role

You will develop and maintain AI agent and RAG pipelines in production, integrate proprietary and open-source language models into compliance and data extraction workflows, build APIs and microservices, and monitor model latency, cost, and accuracy. You will also collaborate on benchmarks, regression tests, vector stores, embeddings, prompt engineering, and context caching.

Requirements

  • Solid experience with Python and backend API development
  • Practical experience with LLMs in production, including LangChain, OpenAI, vLLM, or similar technologies
  • Knowledge of vector databases, embeddings, and semantic retrieval techniques
  • Ability to structure scalable and versioned pipelines using Docker and CI/CD
  • Product mindset and ability to understand the purpose behind what is being built

Responsibilities

  • Develop and maintain agent and RAG pipelines using LangChain, LlamaIndex, or similar frameworks
  • Integrate proprietary and open-source models into compliance and data extraction workflows
  • Build APIs and microservices for document analysis, inconsistency detection, and sensitive information extraction
  • Monitor and optimize model latency, cost, and accuracy in production
  • Collaborate with data and product teams to develop agents for routing and classification decisions
  • Create internal benchmarks and regression tests for LLM and RAG models
  • Participate in technical discussions about vector stores, embedding tuning, prompt engineering, and context caching

Benefits

  • 30 days of paid leave after one year
  • Caju food and transportation allowance for hybrid or onsite employees
  • Wellhub membership
  • Conexa telemedicine
  • Psychologia Viva psychological support
  • Two monthly nutritionist consultations
  • Two monthly psychologist consultations
  • Reduced prices for general practitioners and specialists via telemedicine
  • 15%+ discounts on Dasa network examinations
  • 20% to 30% discounts at Pague Menos pharmacies