We're hiring a senior hands-on AI/ML Engineer to build the core agent stack, from retrieval-augmented generation (RAG) pipelines to multi-tool orchestration and secure, real-time API integrations. This role offers end-to-end ownership and the chance to shape foundational infrastructure at scale.
Key Responsibilities
Design and implement LLM-driven agents capable of multi-step reasoning, contextual memory, and tool use across enterprise APIs.
Build and optimize RAG pipelines, including chunking strategies, semantic retrieval, dynamic prompt construction, and real-time context assembly.
Integrate with organizational platforms (e.g., Slack, Outlook, Jira, Salesforce, Confluence) to provide actionable, personalized, and secure responses. Apply the Model Context Protocol (MCP) to declaratively connect LLMs with tools, memory, and execution components.
Incorporate Agent-to-Agent (A2A) protocols to support multi-agent delegation, communication, and coordination where relevant.
Design with enterprise-grade security in mind, including user authentication, data scoping, redaction, and API-level access controls.
Collaborate closely with AI product and backend engineering peers to ensure performance, reliability, and usability.
36 years of experience in applied ML, NLP, or AI system development.
Strong Python development skills and experience with LLM agent frameworks like LangChain, LlamaIndex, or similar.
Proven experience building agentic systems, including memory management, tool use workflows, and goal-directed execution.
Familiarity with semantic search systems and prompt optimization strategies for retrieval-augmented generation.
Experience working with APIs for platforms like Slack, Jira, Outlook, Salesforce, or Confluence.
Good understanding of secure system design, including permissioning, SSO/OAuth2, and compliance-aware data handling.
Preferred Qualifications
Hands-on experience with Model Context Protocol (MCP) for structured, declarative LLM-to-tool integration.
Experience evaluating agentic and RAG-based systems using task success metrics, tool invocation fidelity, memory consistency, and multi-hop reasoning quality
Exposure to Agent-to-Agent (A2A) protocols for agent coordination, role-based delegation, and inter-agent messaging.
Prior experience with enterprise AI assistants, internal copilots, or intelligent automation tools.
Familiarity with deploying AI systems using Docker, CI/CD pipelines, and cloud infrastructure (AWS, GCP, Azure).
