Description
As the Director of AI Support & Automation, you will build and lead our Bangalore-based engineering pod as an elite 'player-coach' calibrated two levels below the VP. Combining direct architectural leadership with deep, hands-on systems development across MyAgent and Circuit, you will set the benchmark for AI platform reliability and quality.
You will enable the engineering organization to improve and optimize engineering processes and infrastructure, directing software systems development, application engineering, and software quality assurance for cloud-based and agentic AI platforms.
Your directive spans across 5 core pillars:
- Pillar 1: Tier 3 Engineering & Escalated Platform Support: Act as the ultimate engineering escalation authority for MyAgent and Circuit.
Lead hands-on debugging, root-cause forensic analysis, and kernel/runtime patch development for complex production failures.
- Pillar 2: Premier AI Agent Test Harness & Evaluation: Architect and build an end-to-end, agent-driven test automation harness.
Implement automated LLM-as-a-judge scoring frameworks, multi-agent simulation loops, deterministic task verification, and golden dataset benchmarking.
- Pillar 3: AI Software & GPU Performance Validation: Lead performance profiling, stress testing, and concurrency load and latency validation across dedicated on-premise GPU clusters and cloud runtimes.
- Pillar 4: Modular SDK & API Developer Platform: Deliver modular, embeddable SDKs, React components, and headless APIs that empower thousands of Cisco software engineers to control, customize, and extend multimodal agentic experiences.
- Pillar 5: High-Talent-Density Team Leadership: Recruit, mentor, and lead a small, high-density pod (6–10 elite systems, AI, and test automation engineers) in Bangalore.
Technical Architecture & Testing Methodology:
- Deterministic Harness Layer: Unit and integration verification of MCP protocol compliance, tool schema validation, state machines, token exchange, and memory storage persistence.
- LLM-as-a-Judge Evaluation Framework: Automated scoring pipelines using calibrated frontier and on-premise judge models to evaluate grounding, factual accuracy, safety guardrails, hallucination rates, and task completion against versioned golden sets.
- Multi-Agent Simulation & Adversarial Red-Teaming: Automated agent swarms generating synthetic edge cases, malicious prompt injections, tool misuse attempts, and context overflow scenarios to proactively identify failure modes.
- Continuous Canary & Regression Pipeline: Automated staging deployment validation with automated rollback mechanisms, real-time latency tracing, and semantic drift monitoring.
Minimum Qualifications
- Education: Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Applied Mathematics, Artificial Intelligence, or a related quantitative STEM discipline.
- Professional Experience: 10+ years of progressive software engineering experience, including 3+ years leading high-performing technical engineering teams and 2+ years of hands-on production experience with LLMs, agentic workflows, or AI platforms.
- Hands-on Coding Proficiency: Expert-level fluency in Python and modern TypeScript/JavaScript; strong working proficiency in Go, C++, or Rust for systems-level performance tuning.
- AI & Systems Architecture: Deep practical experience with Model Context Protocol (MCP), LangGraph, LlamaIndex, vLLM, Triton Inference Server, PyTorch, and vector databases.
Preferred Qualifications
- Advanced Credentials: Master's degree or Ph.D. in Computer Science, Artificial Intelligence, Distributed Systems, Machine Learning, or Software Engineering.
- GPU & Infrastructure Acumen: Solid understanding of GPU compute infrastructure, CUDA memory management, high-density server configurations, and container orchestration.
- Academic, Research, or Open Source Contributions: Demonstrated academic depth or research contributions in distributed systems, compiler optimization, GPU parallel computing, formal software verification, major contributions to tier-1 open-source AI frameworks/benchmarks, or recognized technical patents in AI/systems engineering.
- Player-Coach Leadership Archetype: Proven ability to lead from the front, write production code, conduct deep architectural reviews, and inspire an elite engineering culture.
- Enterprise Executive Communication: Excellent analytical, written, and verbal communication skills with the ability to influence, interact, and present technical architectures effectively across engineering, product, and executive leadership.