1 день назад
FDE (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
FDE (AI) (LLM agent engineering): Building customer-specific AI skills, integrations, evaluation suites, and speech assets that move from prototypes to stable production with an accent on prompt design, production software engineering, and rigorous evaluation. Focus on integrating customer tools through governed gateways, validating latency in distributed systems, and promoting broadly reusable capabilities into the shared foundation.
Company
provides cloud-based video experiences for organizations, including live, on-demand, and real-time use cases.
What you will do
- Work as a hands-on engineer in a Forward-Deployed Engineering pod embedded with priority customers.
- Build versioned, tested AI skills combining behavioral instructions, executable code, domain knowledge, and customer-specific terminology.
- Wrap and register customer APIs, tools, and MCP servers through the shared gateway with appropriate contracts and permissions.
- Create evaluation suites with ground-truth scenarios and acceptance cases, and validate skills against quality and latency requirements.
- Develop ASR vocabulary biasing and TTS pronunciation lexicons for customer brands, products, and acronyms.
- Feed production learnings back into the platform and promote broadly reusable capabilities into the Foundation.
Requirements
- 4+ years in a technical, customer-facing engineering role with production code responsibility.
- Production experience with LLMs, prompt engineering, agent development, RAG pipelines, tool or function calling, evaluation frameworks, and deployment at scale.
- Ability to author and evaluate AI skills as engineering artifacts, including behavioral instructions, implementation code, test scenarios, and consistency measurements.
- Experience building reusable integrations around external APIs, webhooks, and tools, with an understanding of latency across distributed calls.
- Ability to learn unfamiliar domains, conduct technical discovery, and translate customer conversations into engineering requirements.
- Strong communication, architecture review, delivery planning, prioritization, and blocker-removal skills.
Nice to have
- Production experience with MCP servers, agent frameworks, or multi-agent coordination.
- Background in speech or NLP pipelines, including ASR vocabulary biasing or TTS pronunciation customization.
- Experience with DeepEval, Ragas, Langfuse, Opik, or similar evaluation and observability tools.
- Familiarity with real-time system constraints, multi-tenant platforms, tenant isolation, and registry patterns.
- Experience codifying engineering patterns into tools, playbooks, runbooks, and reproducible benchmarks.
Culture & Benefits
- Hybrid work model with a flexible working approach.
- Collaborative environment where initiative is encouraged and employees have a voice.
- Opportunities to work across customer-specific delivery and shared platform capabilities.
- Exposure to finance, education, media, and telecom domains.
- Growth opportunities across a broad product portfolio and an evolving engineering organization.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →