Job Description
Design, fine-tune, and deploy LLM-based agents using Google ADK, Gemini models, and Vertex AI Agent Builder.
Build autonomous and collaborative agents leveraging A and A (Agents-at-Agents) and MCP (Multi-Context Protocol) frameworks.
Implement RAG pipelines for contextual retrieval from internal knowledge bases, documents, and APIs.
Develop self-improving agent behaviors through reinforcement learning and feedback loops.
Integrate Google LLMs with backend systems (App Engine, Cloud Functions, Big Query, Looker). Orchestrate agent workflows through event-driven architectures and multi-agent communication protocols.
Automate cognitive tasks such as triage, summarization, classification, and reasoning.
Implement evaluation metrics for accuracy, recall, and contextual coherence of agent responses.
Strong experience (8+ years) in applied machine learning, NLP, and LLM fine-tuning.
Hands-on expertise with Google AI stack Gemini Models, Vertex AI, ADK, and Agent Builder.
Proficiency in Python, TypeScript, or Go for developing and orchestrating agents.
Proven experience in RAG pipelines, vector databases (e.g., Pinecone, FAISS, Big Query Vector Search).
Experience with multi-agent orchestration using Lang Chain, Llama Index, or Googles ADK.
Knowledge of MCP (Multi-Context Protocol) and A and A (Agents-at-Agents) collaboration patterns.
Experience in prompt engineering, context grounding, and evaluation frameworks.
Strong understanding of cloud-native architecture, microservices, and API integrations
