RAG AI Engineer - Remote & Flexible Hours

BairesDev; Colombia; Remote; COP 389,395,000 - 584,093,000; Full time

Benefits offered by this job

100% remote

USD compensation available

Home office setup provided

Flexible hours

Paid parental leaves, vacations, andH

Job summary

BairesDev is seeking an AI Engineer specialized in Retrieval-Augmented Generation (RAG) to design and implement advanced RAG pipelines that bind LLMs to external knowledge bases, ensuring grounded and accurate responses.

You will work with Python and vector databases (Pinecone, Qdrant, Weaviate), develop chunking and reranking strategies, and evaluate system performance using established frameworks to deliver production-ready outputs.

Qualifications

• 4+ years of experience in AI Engineering, ML, or Software Engineering.
• Proven expertise in Retrieval-Augmented Generation (RAG) architectures and hybrid search.
• Advanced proficiency in Python and vector databases (Pinecone, Qdrant, Weaviate).
• Hands-on experience with chunking, embeddings, and reranking techniques.
• Familiarity with evaluation frameworks such as RAGAS for model performance.
• Advanced proficiency in English.

Responsibilities

• Design and optimize end-to-end RAG pipelines (ingestion, cleaning, chunking, embedding).
• Implement and manage high-performance vector databases for semantic/hybrid search.
• Develop chunking and reranking strategies to improve precision and context.
• Evaluate and benchmark RAG system performance using established frameworks.
• Collaborate with data and software teams to integrate RAG into enterprise apps/workflows.
• Research and apply techniques like agentic RAG and context compression.

Skills

RAG architectures

Python

Embeddings

Reranking

Evaluation frameworks

English

Tools

Pinecone

Qdrant

Weaviate

Job description

BairesDev is seeking an AI Engineer specialized in Retrieval-Augmented Generation (RAG) to design and implement advanced RAG pipelines that bind LLMs to external knowledge bases, ensuring grounded and accurate responses.

You will work with Python and vector databases (Pinecone, Qdrant, Weaviate), develop chunking and reranking strategies, and evaluate system performance using established frameworks to deliver production-ready outputs.