Director - AI Research and Engineering Efficiency (Circuit Platform)

Cisco•San Jose, CA

3d•$291,500 - $424,400

About The Position

The AI Platforms and Enablement team is the engine behind Cisco's internal AI transformation. Our mission is to build and scale CIRCUIT, Cisco's universal enterprise AI platform designed to collapse information silos and amplify productivity through intelligent, agentic experiences. As a team of researchers, engineers, and AI practitioners, we bridge the gap between powerful industry innovation and practical, production-grade applications. We provide the foundation for Cisco to reimagine work through secure, scalable AI, enabling business units to integrate proprietary data and transform legacy workflows into intelligent, autonomous systems. We value technical depth, rapid prototyping, applied research, production engineering field, and solving complex architectural challenges across models, agents, harnesses, inference efficiency, and enterprise AI governance.

Requirements

Bachelor's degree in Computer Science, Engineering, AI/ML, or a related field, or equivalent practical experience.
15+ years or a PhD with 10+ years of combined experience in software engineering, systems architecture, machine learning, AI platforms or distributed systems.
Experience architecting or operating production LLM, agentic AI, model-serving, or applied AI systems at enterprise scale.
Demonstrated ability to build or lead evaluation systems for LLMs, RAG pipelines, tools, autonomous agents, or model-powered enterprise workflows.
Experience with model efficiency techniques such as model routing, caching, quantization, distillation, prompt/context optimization, batching, or high-throughput inference serving or similar work.
Deep expertise in AI/ML systems, model deployment, distributed platform design, applied AI architecture, and production software engineering practices.
Prior experience driving platform strategy and technical innovation at enterprise scale.
Demonstrated experience influencing product, engineering, security, infrastructure, and leadership organizations through technical strategy, architecture reviews, prototypes, and measurable platform outcomes.

Nice To Haves

Advanced degree or equivalent experience (Master's or PhD) in Computer Science, AI/ML, Systems Engineering, or a related field.
Demonstrated excellence in a Principal Engineer, Distinguished Engineer, Director, or equivalent senior technical leadership role within AI/ML, distributed systems, or enterprise platforms.
Deep expertise in enterprise AI platforms, LLMs, SLMs, agentic systems, and model-serving architectures (e.g., vLLM, TensorRT-LLM, Ray, Kubernetes-based GPU platforms).
Proficiency in the full AI lifecycle, including model training, fine-tuning (LoRA/QLoRA), routing, compression, quantization, and inference optimization.
Extensive experience with agent runtimes, orchestration, tool-calling, human-in-the-loop workflows, memory management, and observability.
Proven ability to build rigorous evaluation frameworks, including golden datasets, adversarial testing, online experimentation, and release gates.
Strong understanding of enterprise AI security, policy enforcement, data privacy, and production-grade governance.
Demonstrated success in delivering complex AI business applications and translating powerful research into scalable, production-grade platform capabilities.

Responsibilities

Direct, mentor, and work alongside a specialized team of AI practitioners.
Contribute to architecture, code reviews, model evaluations, prototype implementations, and technical design decisions to ensure high-quality outcomes.
Define Cisco-scale agentic system architecture, including planning, tool execution, agent-to-agent handoffs, memory, persistent state, human approval, policy enforcement, secure tool access, observability, and failure recovery.
Build rigorous evaluation harnesses for models, prompts, RAG systems, tools, agents, and multi-agent workflows.
Establish offline and online metrics for task success, tool-call accuracy, groundedness, hallucination rate, safety, latency, and cost.
Own the strategy for frontier models, open-weight LLMs, SLMs, fine-tuned domain models, distilled models, and model routers.
Define when Cisco should use vendor-hosted models, self-hosted models, fine-tuned models, or task-specific small models.
Continuously benchmark emerging LLMs, SLMs, and Agent Harnesses.
Lead fine-tuning of open-weight models to meet Cisco's security, domain, latency, cost, and quality requirements.
Drive model-serving efficiency through routing, batching, caching, quantization, distillation, context optimization, structured outputs, and serving-stack improvements.
Own cost-quality-latency tradeoffs across production AI workloads.
Track breakthroughs in AI research and translate them into production-grade capabilities, reusable platform primitives, and validated improvements to CIRCUIT.
Establish technical standards for secure, observable, auditable, policy-compliant AI agents operating over Cisco enterprise data and systems.
Accelerate internal AI development through better tooling, reusable agent patterns, evaluation frameworks, deployment pipelines, and AI-assisted engineering workflows.
Operate at the highest levels of Cisco’s internal AI infrastructure, managing significant platform scale, high-volume inference, and complex enterprise requirements.
Own the optimization of a substantial annual model spend portfolio, drive critical reliability targets (p95/p99 latency, groundedness), and lead a multidisciplinary team of researchers and engineers.
Set the technical AI strategy in partnership with VP/SVP-level leadership across engineering, product, and security.
Be measured on your ability to drive tangible improvements in AI productivity (hours saved, workflow cycle-time reduction), agent reliability (task completion rates, reduction in human overrides), model quality (groundedness, hallucination reduction), inference economics (cost per successful task, GPU utilization), and developer velocity (time-to-production, evaluation coverage).

Benefits

medical, dental and vision insurance
a 401(k) plan with a Cisco matching contribution
paid parental leave
short and long-term disability coverage
basic life insurance
grants of Cisco restricted stock units
10 paid holidays per full calendar year
1 floating holiday for non-exempt employees
1 paid day off for employee’s birthday
paid year-end holiday shutdown
4 paid days off for personal wellness
16 days of paid vacation time per full calendar year (non-exempt)
flexible vacation time off program (exempt)
80 hours of sick time off provided on hire date and each January 1st thereafter
up to 80 hours of unused sick time carried forward
Additional paid time away may be requested to deal with critical or emergency issues for family members
Optional 10 paid days per full calendar year to volunteer
annual bonuses (for non-sales roles)
performance-based incentive pay (for sales roles)