Principal Forward Deployed Engineer GenAI and Telecom Architecture

Location: San Jose CA or Austin TX Or Even Remote

Visa: and EAD

Client: Google (Tech-M)

As a Principal GenAI Forward Deployed Engineer, you are an embedded innovator-builder and principal architect who connects frontier AI products with production telecom operations. You code, debug and jointly ship bespoke agentic solutions inside customer environments while owning the target architecture, integration strategy, production readiness and technical governance required for carrier-grade scale.

Job responsibilities

• Own principal architecture and hands-on engineering for production GenAI and agentic applications, including multi-agent systems, MCP servers and enterprise tool integrations.
• Design target-state and transition architectures connecting Gemini and Vertex AI capabilities with telecom data, network/service APIs, BSS/OSS, CRM, billing, order management, provisioning and customer-care platforms.
• Build structured and unstructured data pipelines, vector retrieval and RAG architectures over telecom knowledge, policy, product, network and customer data.
• Define scalable state-management, evaluation, tracing and observability patterns for distributed agents, tools and customer journeys.
• Establish architecture guardrails for identity, authorization, data residency, privacy, model access, network segmentation, auditability and human approval.
• Lead technical discovery, architecture reviews and production-readiness decisions with customer engineering, network, security, operations and product teams.
• Translate field patterns into reusable reference architectures, modules, accelerators and product feature requests.
• Remain hands-on in Python, APIs, debugging, cloud infrastructure and critical production troubleshooting while mentoring delivery teams.

Minimum qualifications

• Bachelor's degree in Engineering, Computer Science or a related field, or equivalent practical experience.
• At least 8 years of software engineering experience, including 5 years with Python or a comparable language and 3 years in staff, principal or lead solution architecture.
• Production experience architecting AI systems on Google Cloud Platform and shipping customer-facing AI solutions from conception through launch.
• Experience building pipelines for structured and unstructured data using vector databases and RAG architectures.
• Demonstrated telecom-domain architecture involving carrier/customer platforms, BSS/OSS, telecom APIs, subscriber identity, provisioning, billing, network/service operations or telecom-grade contact centers.
• Experience owning integration, security, reliability, scale and operational-readiness decisions across multiple systems and teams.
• Experience leading technical discovery and architecture sessions directly with customers.
• Hands-on experience with Google AI products; direct experience with Gemini, Vertex AI and Google Conversational AI is strongly required for customer deployment readiness.

Preferred qualifications

• Master s or PhD in AI, Computer Science, Telecommunications or a related technical field.
• Production experience with ADK, LangGraph, CrewAI, ReAct, self-reflection, hierarchical delegation or MCP.
• Knowledge of TM Forum APIs and frameworks, event-driven telecom architectures, digital channels, CCaaS/CPaaS, 5G or network/service assurance.
• Experience optimizing LLM-native metrics including tokens per second, cost per request, task success, state management and granular traces.
• Experience designing highly available, fault-tolerant systems for carrier-grade traffic and critical service windows.

Deployment readiness hard stops for both roles

• No demonstrated ownership of enterprise or solution architecture decisions.
• Telecom appears only as a keyword, client name or early-career application project with no architecture depth.
• Candidate cannot describe a telecom system context, integration landscape, security model and production outcome.
• Architecture-only candidate who cannot explain personal coding, deployment or incident-debugging contributions.
• For Requirement 1, no verifiable hands-on Google Conversational AI, CX or CCAI implementation.
• For Requirement 2, no verifiable production GenAI, agentic AI or RAG delivery.