: AI-Native Engineer – Agentic AI & Production Operations

Experience Level: Intermediate / Senior

Job Type: Contract / Full-Time

Role Overview

We are seeking high-performing AI-Native Engineers at Intermediate and Senior levels to design, develop, deploy, operate, and continuously improve enterprise Generative AI, Agentic AI, Multi-Agent systems, and data-centric AI solutions.

This role focuses on engineering and production operations rather than research or prompt engineering alone. Candidates will be responsible for the complete lifecycle of AI solutions, including development, integration, testing, deployment, monitoring, production support, incident resolution, data migration, and continuous evaluation.

The ideal candidate will have hands-on experience with Google Gemini, Google Agent Development Kit (ADK), Vertex AI, BigQuery, Retrieval-Augmented Generation (RAG), vector databases, Python, and Google Cloud Platform (GCP).

Key Responsibilities

- Agentic AI Engineering

- Design, develop, deploy, and maintain production-grade Agentic AI and Multi-Agent systems.
- Implement supervisor/sub-agent orchestration, agent tools, custom extensions, and API integrations.
- Build reliable execution workflows with retries, fallbacks, error handling, and recovery mechanisms.
- Develop and optimize RAG pipelines, vector search, and Human-in-the-Loop (HITL) workflows.
- Troubleshoot agent execution failures, looping behavior, and probabilistic output inconsistencies.
- Maintain and version prompts, configurations, tool definitions, and agent workflows.

2. Production Support and Operations
- Monitor production AI systems and perform regular operational health checks.
- Investigate incidents, analyze logs, perform root cause analysis (RCA), and implement corrective actions.
- Resolve API failures, quota limits, throttling, latency issues, and pipeline failures.
- Manage production deployments, shadow launches, rollback procedures, and configuration updates.
- Implement preventive measures to improve reliability, availability, and operational performance.

3. AI Observability and Monitoring
- Build and maintain observability frameworks for Agentic AI applications.
- Monitor agent latency, invocation counts, task completion rates, failure rates, routing accuracy, and fallback rates.
- Track LLM performance, token consumption, API errors, and quota utilization.
- Detect data quality issues, schema drift, prompt regressions, and model or extraction drift.




- Use monitoring and telemetry tools such as Cloud Monitoring, Cloud Logging, OpenTelemetry, Prometheus, Grafana, or equivalent platforms.

4. Data Engineering and Migration
- Develop and maintain ETL/ELT pipelines, data transformation workflows, and validation scripts.
- Support BigQuery migrations, legacy data modernization, schema changes, and semantic-layer migrations.
- Perform source-to-target data validation, reconciliation, and data integrity checks.
- Manage vector database migrations, RAG knowledge-base synchronization, and metadata updates.
- Implement data retention, archival, and purging policies.

5. RAG and Vector Database Operations
- Maintain ingestion, embedding, indexing, and retrieval pipelines.
- Perform re-indexing when source data changes and validate knowledge-base freshness.
- Troubleshoot retrieval failures, inaccurate context, metadata issues, and ranking anomalies.
- Maintain synchronization between source systems, BigQuery, and vector databases.
- Support automated RAG regression testing and retrieval-quality improvements.

6. Continuous Evaluation and Quality Assurance
- Establish and maintain golden datasets for continuous evaluation and benchmarking.
- Perform shadow-mode testing, side-by-side output validation, and automated prompt evaluations.
- Monitor hallucinations, output quality, behavioral drift, and Human-in-the-Loop acceptance rates.
- Identify issues across data, prompts, models, agent logic, retrieval, tools, infrastructure, and integrations.
- Develop automated testing frameworks to ensure consistent AI application performance.

7. Reliability, Security, and High Availability
- Implement timeouts, idempotency, retry policies, circuit breakers, dead-letter queues, and asynchronous processing.
- Support autoscaling, capacity planning, graceful degradation, and high-availability production services.
- Apply security controls, including IAM, RBAC/ACL, PII masking, secrets management, and audit logging.
- Ensure appropriate data protection, access controls, and compliance with enterprise retention policies.

Required Technical Skills




- AI and Agentic Frameworks: Google Gemini, Google ADK, Vertex AI, LangChain, LlamaIndex, or equivalent agent frameworks.
- Programming: Strong Python development skills, REST/gRPC APIs, debugging, and software engineering fundamentals.
- Cloud and Data Platforms: GCP, BigQuery, ETL/ELT pipelines, data transformation, and schema management.
- RAG and Vector Databases: Vertex AI Vector Search, Pinecone, Chroma, or equivalent vector database technologies.
- Observability: Cloud Monitoring, Cloud Logging, OpenTelemetry, Prometheus, Grafana, Datadog, or equivalent tools.
- DevOps and CI/CD: Docker, Git, automated testing, deployment pipelines, rollback strategies, and configuration versioning.
- Production Engineering: Incident management, RCA, performance optimization, reliability engineering, and continuous evaluation.

Senior-Level Expectations
- Architect complex Multi-Agent systems and production-grade AI solutions.
- Lead incident resolution, root cause analysis, and preventive improvement initiatives.
- Define CI/CD, deployment safety, observability, and continuous evaluation standards.
- Drive migration architecture and resolve complex issues involving latency, quotas, model behavior, and data drift.
- Take end-to-end ownership of system reliability, scalability, and production readiness.

Intermediate-Level Expectations
- Develop, test, deploy, and maintain Agentic AI applications under established engineering standards.
- Support production monitoring, incident triage, debugging, and root cause analysis.
- Maintain RAG pipelines, vector databases, data integrations, and evaluation workflows.
- Implement reliability controls, automated tests, and deployment procedures.
- Collaborate with senior engineers to improve system performance, security, and operational stability.

Idóneo Candidate Profile The ideal candidate combines strong software engineering and cloud data engineering capabilities with hands-on experience operating AI applications in production. Candidates should demonstrate practical experience building Agentic AI workflows, troubleshooting real-world failures, improving observability, and maintaining reliable enterprise AI systems. Important: This is an engineering and production ownership role. Candidates with only theoretical AI knowledge, research-only experience, or prompt engineering experience without hands-on development and production support experience may not be suitable.

📌 AI-Native Engineer – Agentic AI & Production Operations (Guadalajara)
🏢 Centraprise
📍 Guadalajara