AI engineer

Acceler8 Talent; California (MO); On-site; USD 120,000 - 210,000; Full time

Job summary

Acceler8 Talent is partnering with a well-funded enterprise AI company building agents that turn natural-language requests into production-ready data pipelines. The role owns the intelligence layer behind the product, spanning agent architecture, retrieval, code generation, evaluation, and production reliability.

You will design multi-step agents, build evaluation loops for agent quality, develop RAG pipelines with vector search and knowledge graphs, and ensure correctness, observability, and

Qualifications

• 4+ years of software engineering with meaningful AI/ML depth.
• Experience building agents, coding assistants, or generative-AI systems in production.
• Hands-on experience with RAG, retrieval, code generation, or knowledge systems.

Responsibilities

• Own the intelligence layer behind the product, across agent architecture, retrieval, code generation, evaluation, and production reliability.
• Designing multi-step agents for planning, retrieval, generation, and error recovery.
• Building evaluation systems and improvement loops for agent quality.
• Developing RAG pipelines using vector search, knowledge graphs, lineage, and metadata.
• Generating and validating production-grade data transformations.
• Selecting and adapting models for quality, latency, throughput, and cost.
• Taking AI systems from promising prototypes into reliable enterprise production.

Skills

Agent development

Production AI

RAG pipelines

Retrieval systems

Code generation

System design

Tools

Neo4j

AWS

Kubernetes

Databricks

Snowflake

Go

Scala

Job description

I’m working with a well-funded enterprise AI company building agents that turn natural-language requests into production-ready data pipelines.

These agents must understand complex schemas, retrieve the right business context, generate correct transformations, detect errors, and produce workflows that large regulated enterprises can inspect and deploy. The output is working code and data infrastructure, not simply a response in a chat window.

This role will own the intelligence layer behind the product, with scope across agent architecture, retrieval, code generation, evaluation, and production reliability.

You’ll work on problems such as:

• Designing multi-step agents for planning, retrieval, generation, and error recovery
• Building evaluation systems and improvement loops for agent quality
• Developing RAG pipelines using vector search, knowledge graphs, lineage, and metadata
• Generating and validating production-grade data transformations
• Selecting and adapting models for quality, latency, throughput, and cost
• Taking AI systems from promising prototypes into reliable enterprise production

Looking for engineers who have:

• 4+ years of software-engineering experience with meaningful AI/ML depth
• Experience building agents, coding assistants, or generative-AI systems in production
• Hands-on experience with RAG, retrieval, code generation, or knowledge systems

Nice to have:

• Vector-database or knowledge-graph experience, including Neo4j
• Experience with AWS and Kubernetes
• Scala or Go
• Databricks, Snowflake, BigQuery, DuckDB, or Spark experience

This is an opportunity to shape how production AI agents reason over complex enterprise data. You’ll have direct ownership of core architectural decisions and work on systems where correctness, observability, and trust matter as much as model capability.