Senior AI Architect: RAG & Multi-Agent LLM Systems

Aimlroles; Bogotá; On-site; COP 60,000,000 - 120,000,000; Full time

Job summary

Vichara is seeking an experienced AI/ML Engineer to architect and lead multi-agent LLM systems for financial services. You will design RAG pipelines, integrate GPT-4o and PaLM 2, and ensure low latency with robust orchestration.

You will collaborate with DevOps and application teams to deploy on AWS EKS/Azure Kubernetes, implement observability, and drive model evaluation, retrieval quality, and reliability.

Qualifications

• 5+ years as an AI/ML Engineer or similar role.
• Experience with multi-agent AI systems and RAG pipelines.
• Hands-on with LLMs (GPT-4o, PaLM 2) and open-weight models.
• Proficient in MLOps tooling and cloud deployments.

Responsibilities

• Architect and lead multi-agent LLM systems with prompt lifecycle management.
• Build RAG pipelines with hybrid vector search.
• Integrate models and fine-tune transformers for classification and summarization.
• Design and implement agent orchestration and robust fallback behaviors.
• Collaborate across DevOps, MLOps, and UI teams to deliver production-grade AI apps.

Skills

GPT-4o

PaLM 2

LangGraph

LangChain

Promptfoo

SentenceTransformers

LanceDB

Pinecone

Elasticsearch

FAISS

MongoDB

OpenTelemetry

Signoz

Prometheus

Python

SQL

REST APIs

Git

Pandas

NumPy

FastAPI

Docker

Kubernetes

Redis Streams

Azure DevOps

Tools

AWS EKS

Azure Kubernetes Service

Job description

Vichara is seeking an experienced AI/ML Engineer to architect and lead multi-agent LLM systems for financial services. You will design RAG pipelines, integrate GPT-4o and PaLM 2, and ensure low latency with robust orchestration.

You will collaborate with DevOps and application teams to deploy on AWS EKS/Azure Kubernetes, implement observability, and drive model evaluation, retrieval quality, and reliability.