AI Engineer

Entravision

6d•$120,000 - $140,000

About The Position

Entravision is seeking an AI / Applications Engineer to build AI agents and applications that put company data to work. This is an applied engineering role focused on creating and evaluating agents that automate operational workflows and building interfaces through which people can use them. Success is measured by whether agents provide correct, trusted answers in production and whether users adopt the applications. The role emphasizes sound judgment in building reliable LLM systems, with the expectation that the successful candidate can learn Entravision’s Snowflake-based stack. The position is based in Los Angeles, California, and is full time. The compensation range is $120,000 to $140,000. The role reports directly to the Manager, Data Engineering.

Requirements

Strong Python skills, including experience writing production-ready, testable code.
Solid SQL skills and comfort working directly with a data warehouse.
Demonstrated experience building LLM-powered applications that reached production, rather than only prototypes or demos.
Practical understanding of LLM application patterns, including retrieval-augmented generation, embeddings and vector search, prompt engineering, structured output, and grounding responses in governed data.
Experience with agent and tool-calling frameworks such as LangChain, LlamaIndex, MCP, OpenAI or Anthropic SDKs, or comparable tools, and a clear understanding of how to expose data and functions as agent tools.
Experience building user-facing applications with a Python framework such as Streamlit or Gradio, or a JavaScript framework such as React or Next.js.
Experience defining and running evaluations, and ability to reason clearly about failure modes in non-deterministic systems.
Experience with version control, testing, and CI/CD using Git and related DevOps tools.
Bachelor’s degree in Mathematics, Statistics, Computer Science, Engineering, or a related quantitative discipline, or equivalent practical experience.
Strong problem-solving skills and ability to work independently and as part of a small team.
Excellent communication skills, including explaining complex technical concepts to non-technical stakeholders and setting realistic expectations about AI capabilities and limitations.
Interest in staying current with the generative AI and agentic engineering landscape.

Nice To Haves

Experience with vector databases or embedding stores such as pgvector, Pinecone, Weaviate, Qdrant, Elasticsearch, or similar, and semantic search at scale.
Familiarity with LLM observability and evaluation tools such as LangSmith, Langfuse, Braintrust, Arize, or comparable platforms.
Experience with multiple model providers and judgment about model selection, routing, and cost-performance trade-offs.
Experience managing inference cost and latency in production, including caching, batching, and selecting smaller models where suitable.
Familiarity with cloud data warehouses such as Snowflake, BigQuery, Databricks, or Redshift, and their AI-native features.
Experience with containerized deployment using Docker or a managed container service.
Experience with data transformation tools such as dbt.
Awareness of AI governance, including access controls, PII handling, audit trails, and data lineage for AI systems.
A portfolio, repository, or set of shipped projects that the candidate can walk through.

Responsibilities

Design, build, and operationalize AI agents that automate operational workflows across the business.
Develop and register tools and skills that agents invoke, including SQL and function tools, API tools, and search services, and integrate them into agent frameworks and orchestration layers.
Define agent evaluation criteria, build evaluation sets, and measure accuracy, hallucination, latency, and inference cost over time.
Monitor agents in production and use measurements to guide improvements, including detecting regressions when prompts, models, or underlying data change.
Build web interfaces and data applications, including internal tools and agent front-ends, that put data, models, and agents in front of users.
Build retrieval pipelines using chunking, embedding, and indexing strategies to support semantic search and grounding.
Contribute to the context layer used by agents by structuring and enriching trusted data for retrieval-augmented generation and text-to-SQL in partnership with analytics engineering.
Manage tool access and permissions so agents can access only appropriate data.
Partner with business stakeholders to identify workflows suitable for automation, scope them realistically, and set expectations about system capabilities and limitations.
Maintain clear documentation of agent design, tool definitions, prompts, and evaluation results.