Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
$120к-420к/год
$170к$190кмедиана$220к
По похожим ролям (50) ~ $190к/год
На СНГ рынке: $4.3к/мес ($3.7к - $5.4к)
76
Хорошая вакансия
развернуть
Роль дает большой практический опыт в быстро развивающейся области безопасности ИИ, но широкий охват MCP, безопасности, оркестрации и надежности платформы может означать высокую нагрузку.
Широкая зона ответственности
Актуальные технологии
Фокус на ИИ
?Можно работать по контракту
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior MCP / Agentic AI Specialist (Agentic AI): Building and securing production MCP servers, clients, connectors, and agentic workflows that connect large language models with tools, data, and external systems with an accent on protocol implementation, authorization, and agent security. Focus on containing rogue agents, preventing tool-layer attacks, integrating enterprise systems, and ensuring reliable, observable operation at scale.
Location: United States; onshore
Compensation: USD 120,000–420,000, based on experience and qualifications.
Company
hirify.global operates in AI and platform engineering, developing infrastructure for agentic systems and integrations.
What you will do
Build production MCP servers and clients implementing JSON-RPC 2.0, capability negotiation, MCP primitives, tool schemas, and standard transports.
Integrate MCP with Claude, ChatGPT, Cursor, and other hosts, and expose databases, SaaS APIs, and internal services as reliable tools and resources.
Secure the tool layer against tool poisoning, rug pulls, prompt injection, confused-deputy abuse, data exfiltration, and supply-chain risks.
Implement OAuth 2.1, PKCE, resource indicators, audience validation, secrets management, least-privilege authorization, sandboxing, egress controls, and human approvals.
Build agentic and multi-agent workflows using frameworks such as LangGraph, AutoGen, CrewAI, LlamaIndex, and OpenAI or Anthropic agent SDKs.
Improve production reliability through observability, evaluations, health checks, retries, circuit breakers, cost optimization, and technical leadership.
Requirements
6+ years in software engineering, backend or distributed systems, or ML/platform engineering, including hands-on production GenAI or agentic systems experience.
Hands-on MCP experience building servers and clients, designing tool schemas, and working with stdio, Streamable HTTP, legacy SSE, Tools, Resources, Prompts, Sampling, Roots, and Elicitation.
Experience integrating MCP with real hosts and connectors such as Claude, ChatGPT, Apps SDK, or Cursor, and exposing external systems as tools.
Strong understanding of MCP and LLM security, including tool poisoning, rug pulls, prompt injection, confused-deputy and OAuth token issues, and supply-chain risk.
Hands-on OAuth 2.1/OIDC experience with PKCE, resource indicators, audience validation, secrets management, and least-privilege access design.
Experience with LLM applications, function and tool calling, prompt design, RAG, and context or token management.
Familiarity with at least one agent framework: LangGraph, AutoGen, CrewAI, LlamaIndex, OpenAI Agents SDK, or Anthropic agent SDK.
Strong REST, JSON-RPC 2.0, WebSockets/SSE, API integration, and error-handling skills.
Expertise in Python and/or TypeScript; cloud deployment on AWS, GCP, or Azure with containers/Kubernetes and observability.
Nice to have
Open-source contributions to MCP SDKs, the MCP registry, agent frameworks, or MCP security tooling.
Experience with agent guardrails, sandboxing, egress controls, human-in-the-loop approvals, A2A, or multi-agent interoperability.
Experience operating remote MCP servers at scale and familiarity with evaluation tooling, token-cost and latency optimization, OWASP LLM guidance, and MCP security research.
Proficiency in Go or Rust and comfort operating autonomously in a fast-moving domain.
Culture & Benefits
Medical, vision, and dental benefits for full-time employees.
401k retirement plan, variable pay or incentives, paid time off, and paid holidays for full-time employees.
Independent contractor engagement is available.
Hiring process
Applications received more than 120 days after the date of the post will not be considered.
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 →