Senior AI Application Engineer

EVERSANA•New York, NY

7h•$178,000 - $215,000•Hybrid

About The Position

As a Mid-Level AI / Agent Application Engineer, you will be the core builder of the organization's new agentic workforce. Working under the guidance of the Chief AI & Analytics Officer, you will develop the actual agents, write the APIs (tools) the agents will use, and optimize the data pipelines that feed context to the AI.

Requirements

7+ years of software engineering experience, with 3+ years specifically in generative AI, LLMs, or cognitive architectures.
Expert-level proficiency in Python and/or TypeScript.
Experience with MCP architectures
Proficiency in Rust
Deep understanding of agentic design patterns (e.g., ReAct, Plan-and-Solve, Reflection, Tree of Thoughts, etc.).
Extensive experience with LLM APIs (OpenAI, Anthropic, Google Gemini) and open-weights models (Llama 3, Mistral, etc.).
Experience with vector and graph databases
Experience with RAG (Retrieval-Augmented Generation) architectures (e.g., GraphRAG, hybrid search).
Strong background in cloud architecture (AWS, GCP, or Azure) and containerization (Docker/Kubernetes).

Nice To Haves

Ph.D. in Computer Science, Engineering (Electrical, Mechanical, Chemical), Mathematics, Physics, Artificial Intelligence, Software Engineering, or a closely related field.
Experience in enterprise scale deployment of multi-agent architectures.
Expert-level proficiency in Rust.
Extensive experience with RAG (Retrieval-Augmented Generation) architectures (e.g., GraphRAG, hybrid search).

Responsibilities

Architect, design and implement scalable, multi-agent systems that automate complex, multi-step business processes.
Translate existing processes, and develop novel multi-agent architectures for new opportunities
Implement agentic solutions leveraging agent orchestration frameworks (e.g., Google ADK, LangGraph, CrewAI, etc.).
Design secure "tool-calling" architectures, allowing LLMs to interact with internal databases, CRMs, and APIs safely.
Implement LLMOps/AgentOps best practices
Mitigate AI-specific security risks, such as prompt injection, hallucination loops, and unauthorized tool execution.
Demonstrate a commitment to diversity, equity, and inclusion through continuous development, modeling inclusive behaviors, and proactively managing bias.
All other duties as assigned.