Gen AI Engineer III at Elevance Health
Atlanta, GA
—
Jul 23, 2026
Atlanta, GA
Jul 24, 2026
What this job asks for AI summary
A mid-to-senior engineering role focused on building and maintaining complex AI-powered systems, with a particular emphasis on agentic AI, large language model applications, and integrating generative AI tooling into software delivery workflows. Day-to-day work spans prompt optimization, model evaluation, code review, production support, and collaborating with product and QA teams. Best suited to engineers with solid software fundamentals and hands-on experience with AI agent frameworks and cloud-native development.
Mid level · 3+ years · Atlanta-Sandy Springs-Alpharetta, GA · Full-time
“or” means any one of them counts — you don't need all of them.
We read this from the posting text with AI. Skim the description below before ruling yourself out.
How this req sits in the market our data
Roughly 8,100 people in the Atlanta-Sandy Springs-Alpharetta, GA area plausibly meet what this posting asks for (software developers). range 6,100–10,600
Applicant volume Heavy — This req sits in a large pool with little in its requirements to thin it, and auto-apply tools fire at everything in the occupation. Applying early and leading with the rare skills below is what gets read.
What the occupation pays Median $135,884 (middle half $106,941–$167,811).
Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Jul 28, 2026. It is a model, not a headcount.
Why we read it this way (11)
The title 'Gen AI Engineer III' uses a numeric level; 'III' maps to Mid-level advertised seniority, and the minimum 3-year experience requirement also supports Mid.
The role is hybrid (1–2 days/week in-office in Atlanta); it is not fully remote. Alternate locations near another Elevance Health office may be considered.
Visa sponsorship is explicitly not available for this role.
AI coding tools (GitHub Copilot, Cursor, Claude Code, ChatGPT, Codex-style agents) are listed in the responsibilities section with daily/weekly use language — treated as a hard gate on the capability, with the named tools captured as alternatives to each other.
Prompt optimization and model tuning appear in the responsibilities section as core duties; prompt optimization is captured as a required skill. 'Model tuning' is a methodology rather than a named tool and is omitted per guidelines.
Amazon Bedrock AgentCore (including AgentCore Runtime, Memory, Identity, Gateway, Code Interpreter, Browser Tool, Observability, Evaluations, and Policy) is listed under Preferred Skills — captured as Amazon Bedrock and marked preferred.
RAG, memory, model routing, evaluations, guardrails, observability, and secure API/tool access are all listed under Preferred Skills; RAG is captured as a concrete named technique; the others are either methodologies or too generic to emit as discrete named tools.
Healthcare/payer/claims/clinical domain experience (HIPAA, PHI) is listed as preferred.
The degree requirement is a BA/BS 'or any combination of education and experience which would provide an equivalent background' — equivalent experience is accepted, so the degree requirement is set to None.
No compensation figures are provided in the posting.
Ignored 1 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): prompt optimization.
Read the full posting
The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.