K
Forward Deployed Engineer
HybridFull-timeAI EngineeringEnterprise Software
Join a customer-embedded engineering pod to build, evaluate, and deploy production AI capabilities. You will create customer-specific solutions using shared platform components, and promote proven patterns into reusable platform capabilities.
Responsibilities
- Build versioned AI skills that combine behavioral instructions, code, domain knowledge, and customer-specific language.
- Integrate customer tools, APIs, and MCP servers through shared platform components, validating contracts, permissions, and latency.
- Design evaluation suites with domain-specific scenarios and acceptance criteria for deployment gates.
- Create speech assets, including ASR vocabulary lists and TTS pronunciation lexicons.
- Tune deployed skills using customer feedback and production usage.
- Work with customer teams to understand their domains and deliver production-grade solutions.
- Feed platform gaps and reusable patterns back to the foundation team.
Requirements
- At least 4 years in a customer-facing technical engineering role, with production code responsibilities.
- Production experience with LLMs, including prompt engineering, agent development, RAG, tool or function calling, evaluation, and scaled deployment.
- Ability to create and evaluate AI skills using behavioral instructions, code, test scenarios, and consistency measurements.
- Experience building customer-specific integrations that can become reusable, including APIs, webhooks, tools, and distributed-call latency considerations.
- Ability to learn unfamiliar customer domains, conduct discovery, and translate discussions into engineering requirements.
- Clear communication with technical and non-technical stakeholders, including architecture reviews and early risk identification.
- Ability to scope and sequence delivery, remove blockers, and balance scope, speed, and quality.
Nice to have
- Production experience with MCP servers, agent frameworks, or multi-agent coordination.
- Experience with ASR vocabulary adaptation or TTS pronunciation customization.
- Familiarity with LLM evaluation and observability tools such as DeepEval, Ragas, Langfuse, or Opik.
- Experience with real-time systems, multi-tenant platforms, or codifying engineering practices into reusable playbooks and benchmarks.