Client-Facing Agentic AI Engineer

Redapt, Inc.; United States; On-site; USD 120,000 - 190,000; Full time

Job summary

Redapt, Inc. is seeking a Forward Deployed Engineer to turn AI capabilities into production systems within client environments.

You will write production Python code, present findings to client stakeholders, and work across embedding delivery and technical advisory to maximize ROI from AI workloads. You will collaborate with client teams, manage architecture tradeoffs, and optimize token usage, security, and workflow design while traveling to client sites as needed.

Qualifications

• 3+ years in a technical, client-facing role such as forward deployed engineering, solutions engineering, or consulting with a software engineering background.
• Strong Python skills, with experience shipping production code, not just prototypes or notebooks.
• Hands-on production experience with LLMs, including agent development, evaluation, and deployment at scale.

Responsibilities

• Embed directly with client teams to design, build, and ship production applications and integrations built on Claude models.
• Write and maintain production-quality Python code for AI integrations, agent workflows, and internal tooling.
• Set up and configure agentic harnesses so client teams can put AI to work on real business processes.
• Codify repeatable deployment patterns from client engagements and feed them back to the broader team.

Skills

Python

LLMs

Production code

Client-facing

Azure Bedrock

Tools

Azure Foundry

AWS Bedrock

OpenAI/Codex

Claude platform

Job description

Redapt, Inc. is seeking a Forward Deployed Engineer to turn AI capabilities into production systems within client environments.

You will write production Python code, present findings to client stakeholders, and work across embedding delivery and technical advisory to maximize ROI from AI workloads. You will collaborate with client teams, manage architecture tradeoffs, and optimize token usage, security, and workflow design while traveling to client sites as needed.