Agentic AI Developer

Overview As an Agentic AI Developer at Agree Ya, you design, build, and deploy autonomous AI agents and multi-agent systems for enterprise-scale tasks. You’ll translate cutting-edge agent architectures into production-ready solutions, collaborating with product managers, data engineers, and domain experts. You’ll shape reasoning, tool use, memory management, and autonomous execution to solve complex, real-world problems. This role offers impact in advancing intelligent automation within a global services firm. Responsibilities Design and implement autonomous AI agents using modern frameworks (e.g., Lang Graph, CrewAI, Auto Gen, Llama Index, or custom loops)Define agent reasoning architectures (ReAct, Plan-and-Solve, Reflection/Self-Correction, hierarchical multi-agent structures)Build robust tool-use pipelines for external APIs, databases, code execution, and enterprise software Integrate long-term memory and context memory systems; ensure reliable model communication via structured I/O (Pydantic, JSON Schema)Develop fallbacks, HITL checkpoints, and deterministic guards for non-deterministic AI decisions Establish evaluation pipelines (Ragas, Tru Lens, Lang Smith) to benchmark accuracy, speed, and tool reliability; monitor performance, costs, latency, and token usage Debug non-deterministic failures, loops, and hallucinations in multi-step agent execution paths Key requirements Bachelor’s or Master’s in Computer Science, AI, Software Engineering or equivalent with 1–2+ years in LLM/AI applications Expert-level Python; Type Script/Java Script a plus Experience with agent frameworks (Lang Graph, CrewAI, Microsoft Auto Gen, Llama Index Agents, Semantic Kernel)Proven RESTful API development, OpenAPI, function calling architectures Hands-on with vector databases and hybrid retriever architectures (Pinecone, Qdrant, Milvus, Weaviate, Chroma)Strong software engineering fundamentals: asynchronous programming, microservices, Docker/Kubernetes, CI/CD, unit/integration testing for non-deterministic AI systemscollaboration with cross-functional teamsanalytical thinking and problem solvingclear communication and documentation Lang Graph, CrewAI, Microsoft Auto Gen, Llama Index Agents, Semantic Kernel Pydantic, JSON Schema Ragas, Tru Lens, Lang Smith