Senior AI Engineer

Johnson Service Group, Inc.; New York (NY); Hybrid; USD 90,000 - 110,000; Full time

Job summary

Johnson Service Group, Inc. is seeking a contract-focused engineer to design, build, and deploy secure agentic AI systems for production diagnostics, automation, and remediation in New York, NY. The role emphasizes tool-calling agents, model context protocols, and robust security practices.

You will work on RAG pipelines, validator models, and integration with observability and runbook deployment systems to ensure traceability and reliability across global teams.

Qualifications

• 5+ years of software development in Python, C/C++, Go, or Java.
• 3+ years designing, architecting, and launching production ML systems.
• Experience with API integration, prompt engineering, and RAG.
• Experience with vector retrieval, function calling, and secure tool execution.
• Knowledge of OpenAI, Gemini, Llama, Qwen, Claude and related LLM capabilities.
• Strong foundation in statistics, ML concepts, algorithms, and data structures.
• Excellent communication and collaboration across global teams.

Responsibilities

• Design and implement tool-calling agents combining retrieval, reasoning, and secure action execution.
• Implement function calling, orchestration changes, policy enforcement, and least-privilege access.
• Build evaluation frameworks for open-source and foundation LLMs, incl. prompt synthesis, validation, and self-correction loops.
• Integrate agents with observability, incident management, and deployment systems for automated diagnostics and runbook execution.
• Collaborate with production engineers and application teams to define agentic AI roadmaps and measurable objectives.
• Implement validator models, adversarial prompts, policy checks, deterministic fallbacks, circuit breakers, and rollback strategies.
• Build RAG pipelines, curate domain knowledge, validate data quality, and establish feedback loops.

Skills

Python

C/C++

Go

Java

ML systems

RAG

Tool-using agents

Vector retrieval

Secure execution

MCP protocol

OpenAI

Gemini

Llama

Qwen

Claude

Education

Bachelors in CS or equivalent

Tools

Python

C/C++

Go

Java

AWS

Terraform

Job description

Contract opportunity to design, build, and deploy secure, reliable agentic AI systems for production diagnostics, automation, and remediation.

• Location: New York, NY (Hybrid)
• Job type: Contract
• Rate: $65–80 USD per hour (dependent on experience)

Responsibilities

• Design and implement tool-calling agents that combine retrieval, structured reasoning, and secure action execution using Model Context Protocol (MCP).
• Implement function calling, change orchestration, policy enforcement, and least-privilege access controls.
• Build evaluation frameworks for open-source and foundation LLMs, including prompt synthesis, response validation, and self-correction loops.
• Integrate agents with observability, incident management, and deployment systems to support automated diagnostics, runbook execution, remediation, and post-incident summaries with full traceability.
• Collaborate with production engineers and application teams to define agentic AI roadmaps and measurable objectives tied to reliability, risk reduction, and cost.
• Implement validator models, adversarial prompts, policy checks, deterministic fallbacks, circuit breakers, and rollback strategies.
• Build RAG pipelines, curate domain knowledge, validate data quality, and establish feedback loops and milestones to maintain knowledge freshness.

Requirements

• 5+ years of software development experience in one or more languages, including Python, C/C++, Go, or Java.
• 3+ years designing, architecting, testing, and launching production ML systems, including model deployment and serving, evaluation, monitoring, data processing pipelines, and fine‑tuning workflows.
• Practical LLM experience with API integration, prompt engineering, fine‑tuning or adaptation, and applications using retrieval‑augmented generation (RAG) and tool‑using agents.
• Experience with vector retrieval, function calling, and secure tool execution.
• Understanding of commercial and open‑source LLMs and capabilities, including OpenAI, Gemini, Llama, Qwen, and Claude.
• Solid understanding of applied statistics, core ML concepts, algorithms, and data structures.
• Strong analytical problem‑solving, ownership, and urgency, with a focus on measurable business impact.
• Ability to communicate complex ideas clearly and collaborate effectively across global teams.

Technologies

• Python; C/C++; Go; Java
• Machine learning (ML) systems
• LLMs; API integration; prompt engineering; fine‑tuning
• Retrieval‑augmented generation (RAG); tool‑using agents
• Vector retrieval; function calling
• OpenAI; Gemini; Llama; Qwen; Claude
• Model Context Protocol (MCP)

Nice to Have Skills

• Strong hands‑on experience building and maintaining large‑scale Python applications.
• Proficiency building and operating cloud infrastructure, ideally AWS.
• Experience with containerized services using ECS/EKS, serverless computing using Lambda, and data services including S3, DynamoDB, and Redshift.
• Experience with Step Functions, SageMaker, and Infrastructure as Code using Terraform or CloudFormation.