Note: The job is a remote job and is open to candidates in USA. Republic Services is an environmental services company providing recycling, waste, and sustainability solutions. The Staff Engineer - Agentic AI will design, build, and maintain an enterprise-scale agentic AI platform, provide technical leadership, and develop LLM-driven architectures and orchestration pipelines. The role also focuses on LLMOps, AI safety, platform reliability, and mentoring engineering team members.
Responsibilities
- Own the end-to-end development of the enterprise's Agentic AI platform. Design, develop, test, and deploy high-performance generative AI capabilities that allow AI agents to autonomously understand, plan, and execute multi-step tasks with minimal human oversight. Ensure the platform is scalable, highly available, and can support mission-critical applications
- Provide technical direction across the organization on GenAI-related projects. Work closely with Solution and Enterprise Architects to develop solution architectures that integrate LLMs, agent frameworks, and AI services into the broader enterprise system, ensuring alignment with Enterprise Architecture principles and non-functional requirements (security, scalability, resilience, token economics, and latency budgets)
- Lead by example in coding standards, prompt engineering, and context engineering best practices. Conduct code, prompt, and context-pipeline reviews to ensure high code quality, readability, and robust test coverage—including evals, regression suites, and golden datasets for LLM pipelines and API integrations. Establish guidelines for reproducible experiments and version control of prompts, contexts, agents, and datasets (e.g., using Git and LLMOps tools)
- Build internal frameworks and orchestration pipelines to integrate LLMs and agents with enterprise data sources and services. Leverage GenAI tools and protocols (e.g., MCP, A2A, function/tool calling) to enable high-value GenAI business cases across the organization. Design retrieval, memory, and tool-use patterns that make enterprise context reliably available to models at inference time
- Drive continuous improvements in software development and LLM lifecycle processes. Implement LLMOps/GenAI Ops best practices such as automated evaluation, prompt and agent versioning, online/offline evals, observability (traces, token usage, hallucination and grounding metrics), guardrails, and CI/CD pipelines for prompt, agent, and model deployment. Evaluate new tools and methods (e.g., LangSmith, LangFuse, Bedrock, or cloud-based AI services) to enhance team efficiency and system reliability
- Mentor and coach team members in advanced GenAI and software engineering techniques, fostering a culture of knowledge-sharing and innovation. Provide input to management on team performance, hiring, and promotions, helping develop talent in generative AI expertise (non-managerial feedback role)
- Performs other job-related duties as assigned or apparent
Skills
- * 10+ years of experience in designing, developing, and deploying enterprise-scale technology solutions, ideally with two years focused on GenAI/LLM or software architecture initiatives. Demonstrated ability to design and manage complex platforms or products at scale
- * Deep understanding of Generative AI techniques and transformer-based models (e.g., Claude, GPT, or other foundation and open-source LLMs). Hands-on experience integrating, and adapting LLMs into enterprise applications, including expertise in prompt engineering, context engineering, retrieval-augmented generation (RAG), GraphRAG, and agentic patterns (ReAct, planner/executor, multi-agent orchestration) for grounding outputs in enterprise data
- * Familiarity with multi-agent AI systems and agent-based architectures is highly desirable
- * Proficiency in modern GenAI frameworks/libraries such as LangChain, LangGraph, Semantic Kernel, or similar tools. Experience with model-serving runtimes and agent orchestration frameworks for building and managing complex GenAI pipelines. Strong grasp of NLP fundamentals, tokenization, embeddings, and knowledge representation; experience with multimodal models, function calling, structured output, and reinforcement learning from feedback (RLHF/RLAIF) is a plus
- * Solid experience in cloud architectures (AWS Bedrock, kore.ai, GCP AI) for scalable GenAI solution deployment. Familiarity with both containerization and serverless architectures for deploying agents, retrieval services, and model gateways at scale. Understanding of LLMOps/AI DevOps practices (CI/CD for prompts and agents, eval-driven development, prompt and model versioning, observability/tracing, cost and token monitoring, automated red-teaming and guardrail enforcement) to ensure reliable and maintainable GenAI systems
- * Strong background in data engineering and architecture for AI-ready data—skilled in curating, chunking, enriching, and governing unstructured and semi-structured corpora for LLM consumption. Hands-on experience with vector databases (e.g., pgvector, Elastic Search), hybrid search, reranking, knowledge graphs, and embedding strategies for high-quality retrieval. Experience designing semantic layers, metadata, and access controls so enterprise context can be safely surfaced to agents. Experience integrating GenAI solutions via APIs, microservices, and event-driven patterns into existing enterprise platforms
- * Excellent technical leadership and mentorship abilities. Proven track record of guiding engineering teams or projects, conducting technical design reviews, and enforcing best practices. Strong communicator who can translate business requirements into technical solutions and articulate complex GenAI concepts clearly to stakeholders and team members
- * Demonstrated problem-solving skills and a proactive mindset for tackling novel GenAI challenges (hallucination, grounding, long-context reasoning, tool reliability, agent failure modes). Track record of delivering proof-of-concept (POC) projects to evaluate new models, frameworks, and approaches. Stays current with the rapid pace of GenAI research and tooling, and brings innovative ideas (new model architectures, agent patterns, eval methodologies, context engineering techniques) to continuously enhance the platform
- * Deep appreciation for AI ethics, safety, and security considerations in a large enterprise context, including prompt injection, data exfiltration, jailbreaks, PII handling, and responsible AI principles. Experience implementing governance, compliance, and risk mitigation measures for GenAI solutions (e.g., guardrails, content filters, evals for accuracy/bias/toxicity, human-in-the-loop, audit logging). Familiarity with enterprise monitoring, tracing, and logging tools (e.g., Splunk, LangFuse, CloudWatch) to maintain high reliability and site resilience for GenAI services
- * Excellent technical leadership and mentorship abilities. Proven track record of guiding engineering teams or projects, conducting technical design reviews, and enforcing best practices
- * 7 - 10 years of progressively complex experience building and scaling enterprise software systems—combined with recent, hands-on application of GenAI or Agentic AI technologies in production environments
- * Please note, this position is not sponsorship eligible
- * Bachelor's Degree in Computer Science or Data Science - preferred
Benefits
- Target annual bonus of 20% of base salary, subject to individual performance and company performance.
- Comprehensive medical benefits coverage
- Dental plans and vision coverage
- Health care and dependent care spending accounts
- Short- and long-term disability
- Life insurance and accidental death & dismemberment insurance
- Employee and Family Assistance Program (EAP)
- Employee discount programs
- Retirement plan with a generous company match
- Employee Stock Purchase Plan (ESPP)
- Paid Time Off (PTO)
Company Overview