We are seeking a talented and experienced Senior AI Developer to design, implement, and scale out an Agentic AI platform built on Intel CPU/GPU/NPU and edge infrastructure. This role focuses on building a foundation platform that enables both developers and non-technical users to create, orchestrate and deploy AI agents using Intel AI super builder, MCP-based workflows and multimodal AI pipelines. You will work across the full lifecycle of AI systems – from agent creation and orchestration to multimodal processing and business outcome optimization – ensuring solutions are scalable, reliable and production -ready across edge and cloud environments.

This role sits within Lenovo’s Solutions & Services Group (SSG), the global organization that brings together our end‑to‑end AI solutions and services to turn customer vision into value. You'll be joining a new distributed engineering team which extends Personal agentic platforms like OpenCLAW, Hermes etc. into AI native personal delivery platforms that powers Lenovo's Agentic AI strategy across Hybrid cloud, on-Prem and edge.

Key Responsibilities:

Platform and Architecture

• Design and develop a scalable Agentic AI platform leveraging Intel AI super builder and MCP based architectures
• Build end to end systems including Agent frameworks, orchestration layers, inference pipelines and user interfaces
• Integrate AI solutions across edge, on-premise and cloud environments

Agent Frameworks and workflow orchestration

• Develop and Extend MCP-based work flow orchestration and Data ingestion systems
• Implement Trigger-based workflows ( Scheduled RAG + MCP pipelines)
• Multi-agent Coordination pipelines ( eg : evaluate à summarize à score)
• Seamless integration of multiple data source and MCP servers
• Build reusable agent templates and tools

Multimodal AI systems

• Develop multi modal AI pipelines for
• • Audio ( transcription, summarization, chapterization)
  • Video ( CCTV/Ring-style summarization, event detection)
  • Vision-language reasoning (VLM-based systems)
• Implement orchestration and validation pipelines for multimodal workflows
• Optimize inference for edge deployment on Intel CPU/GPU/NPU platforms

Business-oriented Agents

• Build domain-specific AI agents aligned to measurable business outcomes, including
• • Staff interaction tracking
  • Customer journey intelligence
• Translate AI outputs into business KPIs and actionable insights
• Collaborate with stake holders to align AI systems with revenue, efficiency and customer experience goals

Control plane, GenAIOps and performance optimization

• Design and implement a centralized control plane for Agent orchestration, routing scheduling and life cycle management
• Design and reliability agent for monitoring, evaluation, guardrails
• Maximize utilization across Intel CPU/GPU/NPU and dynamically schedule workloads

Basic Qualifications:

• Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, or a related field.
• 5+ years in AI/ML system development across domains (NLP, CV, Multimodal AI)

Preferred Qualifications:

• 2+ years of experience in developing Generative AI / Agentic AI systems in production environments
• Strong experience with
• • LLMs, SLMS, VLMS
  • Agent orchestration frameworks ( MCP or similar paradigms)
• Experience building multi-agent systems and workflow pipelines
• Knowledge of RAG architectures and vector databases

Systems and Infrastructure

• Experience with Edge + on prem + cloud deployments
• Kubernetes and containerized environments
• Familiarity with Intel AI stack ( Open Vino, CPU/GPU optimization) is a plus
• Experience with NVIDIA stack ( CUDA, TensorRT) is a plus

Programming

• Strong coding skills in Python ( Mandatory)
• Frontend frameworks ( React or similar)
• Scripting and automation tools

What Makes this role Unique

• Opportunity to build production -grade Agentic AI platform on Intel Hardware
• Work across
• • Agent orchestration ( MCP)
  • Multimodal AI systems
  • Edge-to-cloud deployment architectures
• Direct impact on business KPI-driven applications in real -world environments