Senior Agentic AI Engineer

Adobe Adobe · Enterprise · Yerevan, Armenia

Senior Engineer to build AI-native platforms and agentic AI systems that integrate LLMs, recommendation engines, and multi-agent orchestration for personalized digital experiences at Adobe's global scale. The role involves designing, implementing, and prototyping these systems, with a focus on production readiness, evaluation, and responsible AI guardrails. Experience with LLM APIs, multi-agent frameworks, RAG, vector databases, and distributed systems is required.

What you'd actually do

  1. Contribute to building AI-native systems where agents, LLMs, and recommendation engines collaborate to deliver personalized digital experiences on a global scale.
  2. Design and implement agentic AI systems that combine LLM reasoning, tools, structured data, and multi-agent orchestration (A2A, MCP) to deliver adaptive customer experiences.
  3. Create and contribute to engineering designs and architectures that address real product and platform challenges at the Adobe scale.
  4. Prototype rapidly, run PoCs, experiment with emerging AI capabilities, and iterate based on measurable impact.
  5. Implement evaluation, monitoring, and Responsible AI guardrails to ensure reliability, fairness, and production readiness.

Skills

Required

  • Python
  • high-availability services
  • distributed systems
  • scalable architectures
  • data modeling
  • LLM APIs (OpenAI, Anthropic, LiteLLM)
  • streaming
  • tool calling
  • structured outputs
  • multi-agent orchestration frameworks (LangGraph, CrewAI, Agno, or similar)
  • A2A
  • MCP
  • RAG architectures
  • vector databases
  • semantic retrieval systems
  • cloud-native infrastructure (Docker, Kubernetes, AWS/GCP/Azure)
  • distributed data platforms (Kafka, Airflow, Snowflake, SingleStore, or similar)
  • evaluation of metrics (NDCG, MAP, precision/recall)
  • online A/B testing frameworks
  • ranking strategies
  • experimentation methodologies
  • relevance optimization

Nice to have

  • JavaScript
  • Java

What the JD emphasized

  • building and operating scalable distributed systems in production environments
  • multi-agent orchestration frameworks
  • Responsible AI guardrails

Other signals

  • building AI-native platforms
  • agents, LLMs, and recommendation engines collaborate
  • global scale
  • multi-agent orchestration