Azure

NewRocket

23h•Remote

About The Position

NewRocket is seeking a Senior Forward Deployed AI Engineer to join its AI Foundry team and work directly with customers to deploy, operationalize, and scale AI-powered workflow solutions. The role combines full-stack engineering, enterprise integration, generative AI implementation, and client-facing solution delivery. The engineer partners with consultants, product teams, AI/ML engineers, and customer stakeholders to turn business problems into secure, reliable, production-ready AI solutions. The role involves delivering AI solutions using Claude, Anthropic APIs, Microsoft Copilot Studio, Azure AI Foundry, and other appropriate AI technologies. Work includes prompt and context engineering, retrieval-augmented generation, tool use, structured outputs, agentic workflows, model evaluation, and responsible AI controls. The engineer will support AI assistants, automations, and integrations in ServiceNow and wider enterprise ecosystems, and contribute to NewRocket platforms and reusable assets, including the NewRocket Intelligence Platform, Value Realization Dashboard, Data Intelligence Platform, and Agent Packs. This is a customer-facing role requiring sound engineering judgment, curiosity about emerging AI technologies, responsible deployment practices, and the ability to work in fast-moving customer environments. The position is remote with approximately 70% travel.

Requirements

At least 8 years of experience in software engineering, systems integration, enterprise platforms, workflow automation, or related technical delivery roles.
Strong engineering foundation with hands-on full-stack development, scripting, APIs, microservices, enterprise integrations, or cloud-native applications.
Experience building, deploying, or supporting AI/LLM-powered applications, AI-enabled automations, conversational experiences, RAG systems, or agentic workflows.
Experience integrating APIs, enterprise applications, data platforms, or workflow systems in production environments.
Experience with Power Automate and Power BI.
Proficiency in JavaScript/TypeScript, Python, or similar programming and scripting languages.
Familiarity with prompt engineering, context windows, token usage, embeddings, vector search, RAG, tool use/function calling, structured outputs, and model evaluation.
Understanding of responsible AI, including hallucination mitigation, sensitive-data handling, identity and access controls, human oversight, AI safety, and secure AI deployment.
Experience in customer-facing engineering, consulting, technical implementation, solutions architecture, or professional services.
Strong problem-solving skills and ability to communicate complex technical concepts to technical and business stakeholders.
Ability to manage ambiguity, prioritize effectively, travel to clients as needed, and deliver quality solutions in fast-moving customer environments.

Nice To Haves

Familiarity with Model Context Protocol (MCP) and experience building or integrating secure tools and data connections for AI applications.
ServiceNow development, configuration, workflow automation, or enterprise platform implementation experience.
Familiarity with ServiceNow scripting, APIs, IntegrationHub, Flow Designer, Virtual Agent, Now Assist, AI Agents, and platform development patterns.
Understanding of ServiceNow data models and enterprise workflow concepts, including CMDB, ITSM, CSM, HRSD, employee workflows, workflow and task tables, knowledge management, and enterprise integration and data exchange patterns.
Experience building AI prototypes, RAG systems, semantic search capabilities, AI assistants, or agent-based workflows.
Experience integrating LLM APIs or AI services into enterprise applications and business processes.
Experience with vector databases, embeddings, document ingestion, chunking, retrieval strategies, and knowledge-grounding patterns.
Familiarity with frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, or comparable tools.
Experience with AI observability, tracing, evaluation frameworks, prompt/version management, monitoring, and cost optimization.
Experience contributing to internal platforms, reusable accelerators, developer tooling, or product capabilities.
Familiarity with Docker, CI/CD, SQL, Git, infrastructure-as-code, and secure cloud deployment practices.

Responsibilities

Deploy, configure, and operationalize agentic AI workflows, assistants, and automations in customer ServiceNow environments and enterprise ecosystems.
Translate business requirements, processes, and outcomes into technical architectures, implementation plans, and production-ready AI solutions.
Implement NewRocket Agent Packs, AI accelerators, and workflow solutions in enterprise environments.
Tailor AI solutions to customer operating models, processes, data sources, security requirements, and user needs.
Support customer workshops, discovery sessions, technical working sessions, demonstrations, pilots, and production rollouts.
Explain AI capabilities, limitations, tradeoffs, solution behavior, and adoption considerations to technical and business stakeholders.
Build enterprise AI applications and workflows using Claude, the Anthropic API, and other LLM platforms as appropriate.
Apply prompt and context-engineering techniques, including structured inputs, response constraints, and long-context management.
Design AI workflows using structured outputs, tool use, API integrations, multi-step orchestration, and human review.
Build RAG solutions grounded in authorized enterprise data and knowledge sources.
Improve AI reliability and trust through source grounding, validation, confidence thresholds, schemas, fallback handling, and escalation workflows.
Stay current with Anthropic platform capabilities, Claude releases, implementation guidance, responsible AI principles, and enterprise deployment practices; complete relevant partner training and technical education.
Build demonstrations, prototypes, and proofs of concept to validate AI workflows and customer use cases.
Iterate on solutions with customers and internal teams based on feedback, performance, and operational needs.
Support the design and implementation of AI orchestration, LLM integrations, agentic decision models, and workflow automation patterns.
Assess whether agentic AI or deterministic automation, traditional workflow logic, search, analytics, or human review is appropriate for a use case.
Develop reusable implementation patterns, templates, prompt libraries, integrations, and deployment assets.
Develop secure integrations between ServiceNow, enterprise systems, APIs, data platforms, and AI services.
Build supporting scripts, microservices, automation logic, integration services, and lightweight user interfaces.
Integrate AI platforms with enterprise APIs, identity systems, document repositories, databases, and structured and unstructured data sources.
Apply engineering practices for authentication, authorization, secrets management, access controls, logging, error handling, version control, and documentation.
Design solutions for enterprise security, scalability, maintainability, observability, and production readiness.
Develop and execute AI evaluation approaches using test cases, representative datasets, success metrics, and regression testing.
Assess AI workflow quality for relevance, accuracy, groundedness, task completion, safety, latency, cost, and user experience.
Implement safeguards for sensitive data, role-based permissions, appropriate data access, prompt injection, unsafe tool use, and unintended model behavior.
Establish human-in-the-loop workflows for sensitive, high-impact, low-confidence, or exception-based decisions.
Document solution behavior, limitations, risk controls, governance considerations, and support procedures.
Monitor and improve deployed solutions using user feedback, usage patterns, performance data, incidents, and changing customer needs.
Contribute reusable assets, integrations, prompt patterns, evaluation frameworks, and automation capabilities to NewRocket platforms and intellectual property.
Identify common patterns, requirements, integration needs, and capabilities discovered through customer deployments.
Provide product feedback and help convert successful client implementations into repeatable platform features, accelerators, and delivery playbooks.
Support standards and best practices for enterprise AI delivery across the AI Foundry.
Diagnose and resolve technical issues involving AI workflows, integrations, retrieval pipelines, data connections, automation, model inputs and outputs, prompts, tools, APIs, permissions, data quality, and performance.
Optimize deployed systems for reliability, performance, latency, model usage, and cost efficiency.
Collaborate with Business Process Consultants, Product Engineering, Data Engineers, AI/ML Engineers, ServiceNow teams, and the AI Center of Excellence.
Serve as the engineering counterpart to consulting and delivery teams across discovery, design, implementation, rollout, and continuous improvement.
Contribute to internal playbooks, technical documentation, deployment patterns, reference architectures, and product evolution.
Design, develop, and deploy autonomous and human-in-the-loop agents using Microsoft Copilot Studio, Azure AI Foundry, and related Microsoft technologies.
Build agentic workflows that reason, plan, execute actions, and interact with enterprise applications.
Develop multi-agent solutions with orchestration, tool calling, contextual memory, and workflow coordination.
Implement RAG using Azure AI Search, vector databases, and enterprise knowledge sources.
Extend Copilot capabilities through custom connectors, APIs, plugins, and MCP integrations.
Develop custom agent capabilities using Python, C#, TypeScript, and relevant AI orchestration frameworks.
Integrate Jira, Confluence, and Smartsheet.
Use Azure AI Foundry for model deployment, agent development, evaluation, and operational management.
Achieve the stated first-six-month outcomes, including deployments across customer engagements, trusted customer relationships, practical use of Claude and LLM engineering practices, reusable contributions, customer-driven product feedback, and repeatable methods for governed production deployment.