About the Company
AvantoDev is dedicated to building innovative solutions that empower agents through data-driven insights. Our mission is to create a seamless and efficient data ecosystem that enhances decision-making and operational efficiency.
About the Role
The Senior Data Engineer owns the data backbone of AvantoDev's agentic platform — the pipelines, schemas, and stores that feed every agent, MCP server, and RAG retrieval path in the "Strata" Operating Layer. Our agents are only as good as the context they receive, and that context lives in your pipelines: tenant-specific master data, document extraction outputs, schema registries, vector embeddings, and the immutable audit trails that make nondeterministic systems governable.
Responsibilities
- This is not a "wire up a dbt job" role. You will architect the Bronze → Silver → Gold data lifecycle that turns high-entropy inputs (PDFs, SIF files, ACKs, POs, Invoices) into clean, validated, agent-consumable data.
- You will design the PostgreSQL/pgvector schema registry,
build and operate the OpenSearch state-tracking and analytics layer, run the embedding and retrieval pipelines behind our RAG strategies, and ensure that every byte that reaches an agent is traceable, tenant-isolated, and cost-accounted.
- You will work across the full data stack — from Aurora Serverless v2 and pgvector, to OpenSearch index design, S3 data lakes, SQS/Step Functions orchestration, and Terraform infrastructure — collaborating closely with the CIO (Leonardo Vargas), the AI Engineering team, and the Head PM.
Qualifications
- 5+ years building production data pipelines, with strong Python (async patterns, type hints, modern Python 3.10 and advanced SQL
- Hands-on with vector databases / RAG data plumbing — specifically pgvector IVFFlat vs. HNSW, embedding pipelines, chunking strategies, and hybrid search (keyword + semantic, RRF.
- Deep PostgreSQL expertise — schema design, migrations, query optimization, indexing, and partitioning
📌 Senior Data Engineer (Bogotá)
🏢 Avanto
📍 Bogotá