The forward-deployed AI Engineer designs, builds, and ships AI-powered solutions, including agentic systems, that transform workflows and unlock new capabilities across NBCUniversal News Group. This role is deeply technical, highly collaborative, and execution-focused, with success measured by production adoption and measurable operational impact.
This individual works end-to-end from discovery and prototyping through deployment and iteration, combining LLMs, retrieval, data integration, and orchestration frameworks to deliver reliable solutions in production. This is not a backend-only role; the engineer will build across services, integrations, and user-facing experiences as needed. Engineers will work closely with editorial, product, data, and technology teams to co-create solutions in a fast-moving, real-time business environment where speed and timely delivery are critical.
Key Responsibilities:
Design, Build, and Ship AI + Agentic Solutions
- Architect and implement end-to-end AI applications from concept to production, including agentic workflows.
- Build systems using LLMs and multimodal models, retrieval and vector databases, and orchestration frameworks (e.g., LangChain/LangGraph, Google ADK), integrated with internal tools and data.
- Own the full lifecycle: discovery, prototyping, validation, deployment, monitoring, iteration, and support.
Execute Fast in a High-Velocity Environment
- Deliver solutions on accelerated timelines aligned to editorial and business urgency.
- Prioritize evidence-driven iteration over perfection; make clear tradeoffs across scope, speed, and quality.
- Adapt quickly to changing requirements and real-world feedback, keeping stakeholders aligned.
Hands-On Technical Execution
- Write production-quality code and integrate APIs across internal platforms and external providers (e.g., Google, OpenAI, Databricks, AWS, Microsoft).
- Make pragmatic architectural decisions balancing speed, scale, security, reliability, and cost.
- Build reusable components, templates, and playbooks that accelerate future development and reduce friction for downstream teams.
Apply the Right AI to the Right Problems
- Evaluate model capabilities and select appropriate AI tools based on workflow needs, risk profile, and operational constraints.
- Combine services across selected providers into cohesive, reliable systems with strong observability and clear failure modes.
- Implement evaluation frameworks to measure quality, reliability, and impact, and to guide iteration.
Responsible AI, Safety, and Governance
- Design guardrails for agentic systems, including tool safety, permissioning, data handling, and human-in-the-loop review where needed.
- Partner with Standards, Legal, Security, and GRC to ensure compliant data use, responsible deployment, and audit-ready practices.
- Establish monitoring and incident response patterns to detect issues and maintain trust and reliability at scale.
Collaborate Deeply Across Teams
- Work directly with editorial, product, and business partners to shape solutions from requirements through rollout.
- Collaborate with our Data team on data quality, retrieval strategies, evaluation design, and measurement.
- Partner with the CTO and platform engineering organizations to harden solutions for production environments.
- Communicate clearly with technical and non-technical audiences, translating complexity into decisions and action.
Stay Ahead of the Market
- Track new model releases, vendor updates, and emerging patterns in agentic AI.
- Run targeted experiments with new capabilities and share timely recommendations that influence roadmaps and standards.
Leverage AI in the Role
- Use AI coding agents and copilots to accelerate development, debugging, refactoring, test generation, and documentation.
- Apply AI to speed analysis, coordination, and knowledge capture while maintaining security, quality, and compliance.