Implement agentic memory and outbound MCP servers exposing retrieval, graph traversal, and lineage to external agents with access controls gatekeeping and full audit.
Engineer governance so access control propagates from source data through embeddings, graph nodes, retrievals, and agent responses.
Benchmark retrieval quality, embedding models, and LLMs against EDA use cases, selecting models per task under cost and latency constraints.
Collaborate with product, EDA tool teams and customers to translate semiconductor and RF workflows into requirements.
Qualifications
MS or PhD in Computer Science, Electrical Engineering, or related field
5+ years building production ML or data-intensive systems.
Demonstrated experience building RAG and knowledge graph systems in production (a must): ingestion, indexing, and retrieval pipelines.
Hands-on expertise with LLMs: embeddings, fine-tuning, prompt and context engineering, evaluation, and open-weights models for on-prem inference.
Strong command of vector databases, graph databases, and low-latency retrieval infrastructure at scale.
Experience with agentic memory management and the Model Context Protocol (MCP) or comparable agent-grounding interfaces.
Proficiency in Python and modern ML frameworks (PyTorch, Hugging Face, etc..), with solid software engineering and API design practices.
ML applied to semiconductor applications, especially the RF and microwave industry, is highly valued.
Familiarity with data governance, access control, and provenance in IP-sensitive or regulated environments is a plus.
Careers Privacy Statement***Keysight is an Equal Opportunity Employer.***
Design and build low-latency hybrid retrieval (lexical, vector, graph, faceted) over large, heterogeneous data for on-premises, IP-sensitive deployments.
Develop the semantic insight layer: automated tagging, domain-aware metadata, and embeddings as first-class managed assets with version provenance.
Build the EDA-aware knowledge graph as organizational memory: entity and relationship inference, ontology evolution, versioning, and temporal queries.
Implement agentic memory and outbound MCP servers exposing retrieval, graph traversal, and lineage to external agents with access controls gatekeeping and full audit.
Engineer governance so access control propagates from source data through embeddings, graph nodes, retrievals, and agent responses.
Benchmark retrieval quality, embedding models, and LLMs against EDA use cases, selecting models per task under cost and latency constraints.
Collaborate with product, EDA tool teams and customers to translate semiconductor and RF workflows into requirements.
MS or PhD in Computer Science, Electrical Engineering, or related field
5+ years building production ML or data-intensive systems.
Demonstrated experience building RAG and knowledge graph systems in production (a must): ingestion, indexing, and retrieval pipelines.
Hands-on expertise with LLMs: embeddings, fine-tuning, prompt and context engineering, evaluation, and open-weights models for on-prem inference.
Strong command of vector databases, graph databases, and low-latency retrieval infrastructure at scale.
Experience with agentic memory management and the Model Context Protocol (MCP) or comparable agent-grounding interfaces.
Proficiency in Python and modern ML frameworks (PyTorch, Hugging Face, etc..), with solid software engineering and API design practices.
ML applied to semiconductor applications, especially the RF and microwave industry, is highly valued.
Familiarity with data governance, access control, and provenance in IP-sensitive or regulated environments is a plus.
Careers Privacy Statement***Keysight is an Equal Opportunity Employer.***
