Full Stack AI Engineer (Conversational & Voice AI)

Quick Facts

• Role: Full Stack AI Engineer II (Conversational AI & Voice AI)
• Contract, Remote (United States)
• Interview: Video

Description

The Full Stack AI Engineer II designs, builds, tests, deploys, and operates enterprise-grade AI agents across digital and voice channels. The role combines full stack engineering, LLM application development, conversational design, cloud-native delivery, security, and production operations to translate business needs into dependable agent experiences integrated with enterprise APIs and governed data.

Responsibilities

• Build and maintain production LLM agents using prompt design, structured outputs, tool calling, retrieval patterns, workflow orchestration, and multi-agent coordination
• Develop conversational solutions using Google Customer Engagement Suite Agent Studio, Dialogflow CX, and Agent Development Kit (ADK) (intents, flows, tools, webhooks, environment configuration)
• Design low-latency voice experiences (turn-taking optimization, barge-in, graceful recovery from speech recognition errors, silence/retry handling)
• Develop Python and TypeScript/Node.js services, APIs, integrations, orchestration components, and lightweight interfaces for user/operations
• Integrate agents with internal and external APIs, event-driven services, business systems, and governed data sources with resilient authentication, error handling, retries, idempotency, and rate-limit controls
• Create evaluation assets (conversational regression suites, simulation-based tests, golden conversation sets, quality rubrics, release gates) measuring task completion, correctness, safety, latency, escalation quality, and user outcomes
• Implement agent identity using Microsoft Entra Agent ID and standards-based OAuth 2.0/OIDC patterns with token exchange, scoped authorization, secure secret handling, and least-privilege access
• Build and operate Google Cloud Platform services using Cloud Run, IAM, Secret Manager, BigQuery, logging, monitoring, and related managed services
• Design and operate production RAG pipelines (ingestion, chunking, embeddings, vector/hybrid search, metadata filtering, re-ranking, grounding, retrieval evaluation)
• Use Terraform for repeatable cloud infrastructure, service configuration, access policies, and environment setup
• Establish CI/CD practices for prompts, tools, flows, models, and configuration across development, test, staging, and production environments
• Implement observability and analytics (structured logging, distributed tracing, conversation analytics, auditability, dashboards for containment, transfer rate, outcomes, quality, errors, cost, and latency)
• Use GitHub Copilot/Copilot custom agents and skills with code review, testing, provenance checks, and human-in-the-loop guardrails
• Support design reviews, incident response, root-cause analysis, performance tuning, documentation, and continuous improvement of reusable AI engineering patterns

Required Qualifications

• Bachelor’s degree in a related field or equivalent practical experience
• 3+ years of professional software engineering; hands-on backend/full stack development with Python and TypeScript/Node.js
• 1+ year building LLM-powered applications, conversational AI solutions, or production automation using prompts, tools, APIs, and evaluation methods
• Experience designing RESTful APIs, asynchronous workflows, service integrations, and cloud-native applications
• Working knowledge of OAuth 2.0, OIDC, service identities, token handling, secrets management, and least-privilege security practices
• Experience with CI/CD, Git-based development, automated testing, containerized deployment, and infrastructure as code
• Ability to troubleshoot distributed systems using logs, traces, metrics, and reproducible test cases
• Strong written and verbal communication; cross-functional collaboration

Preferred Qualifications

• Experience with Dialogflow CX, Google Customer Engagement Suite Agent Studio, and Google Agent Development Kit (ADK)
• Experience with Cloud Run, Google Cloud IAM, Secret Manager, BigQuery, Cloud Logging, and Cloud Trace
• Experience building real-time voice agents with streaming speech-to-text, text-to-speech, telephony integration, interruption handling, and low-latency orchestration
• Experience with Microsoft Entra Agent ID and workload identity patterns
• Experience with multi-agent systems, agent-to-agent interactions, tool registries, or Model Context Protocol integrations
• Experience with offline/online evaluation pipelines, conversation simulators, regression suites, and automated release gates for AI systems
• Experience with Terraform modules, policy controls, environment promotion, and ephemeral preview/branch environments
• Familiarity with responsible AI practices (prompt injection defenses, content safety controls, privacy-by-design, human escalation patterns)

Technical Competencies

• LLM and agent engineering: prompting, tool calling, structured outputs, orchestration, evaluations, guardrails, release management
• Conversational and voice AI: Dialogflow CX, Agent Studio, ADK; turn latency, barge-in, recognition recovery, containment
• Application engineering: Python, TypeScript/Node.js, APIs, event-driven integration, testing, secure coding
• Cloud and platform: Google Cloud Platform (Cloud Run, IAM, Secret Manager, BigQuery, containers, Terraform)
• Identity and security: Entra Agent ID, OAuth 2.0, OIDC, token exchange, secrets, least privilege
• DevOps and operations: Git, CI/CD, environment promotion, branch agents, logging, tracing, analytics, incident support