We're looking for a highly skilled and visionary Agentic AI Implementation Engineer who can design, build, and deploy intelligent systems using Agentic RAG (Retrieval Augmented Generation) techniques and custom GPTs. This pivotal role will involve creating autonomous or semi-autonomous agents powered by Large Language Models (LLMs) that are capable of sophisticated task planning, contextual decision-making, and iterative information retrieval. You will operate at the intersection of advanced prompt engineering, tool orchestration, and multi-step reasoning, with a strong focus on performance, scalability, and user-centric design.
Key Responsibilities
- Architect & Implement Agentic Workflows: Design and implement robust agentic workflows leveraging RAG pipelines, LLM agents, and external tool integrations.
- Design Modular Systems: Create modular, agentic systems that seamlessly incorporate planning, memory, tool use, and context-aware reasoning capabilities.
- Develop & Optimize Custom GPTs: Develop and fine-tune custom GPTs utilizing advanced prompt engineering and OpenAI's custom instructions, functions, and APIs.
- Integrate Knowledge Bases: Integrate diverse knowledge bases, vector stores (e.g., FAISS, Pinecone, Weaviate, Cosmos DB, ChromaDB), and APIs into a cohesive Agentic RAG architecture.
- Fine-tune Agent Behaviors: Fine-tune agent behaviors for a variety of real-world applications, such as customer support, research assistants, and code agents.
- Collaborate Cross-functionally: Work closely with product managers, UX designers, and backend engineers to ship scalable and robust AI solutions.
- Rapid Prototyping & Experimentation: Rapidly prototype ideas, conduct LLM experiments, and iterate on designs using both quantitative and qualitative metrics.
- Monitor & Optimize Performance: Continuously monitor system performance, detect and address reasoning failures, hallucinations, and retrieval mismatches to ensure high-quality outputs.
- Stay Updated: Remain current with the latest research and advancements in Agentic AI, RAG, tool use, and autonomous agents.
Must-Have Skills
- Experience: Proven experience in software engineering, ML systems, or applied NLP/LLM development.
- Agentic RAG Frameworks: Strong expertise in Agentic RAG frameworks such as LangGraph, AutoGPT, CrewAI, and LangChain Agents.
- Custom GPTs & OpenAI API: Demonstrated ability to design and implement custom GPTs using advanced prompt strategies and the OpenAI API (functions, tools, memory).
- Vector Databases & Embeddings: Hands-on experience with vector databases (e.g., Cosmos DB, Pinecone, ChromaDB), embeddings, and semantic search.
- Retrieval Augmentation: Deep understanding of retrieval augmentation, context compression, multi-hop querying, and memory management.
- Programming & LLM Tooling: Fluency in Python and experience with modern LLM tooling (e.g., LangChain, LlamaIndex).
- Systems Thinking: Strong systems thinking and the ability to effectively balance trade-offs between model performance, latency, and accuracy.
- Agile Environment: Comfortable with fast-paced, iterative environments and exploratory development.
Desired Skills
- Autonomous Agents: Experience with autonomous agents and frameworks like AutoGen, OpenAgents, or BabyAGI.
- AI Safety & Ethics: Understanding of AI safety, ethics, and control mechanisms in agentic systems.
- LLM Evaluation: Familiarity with evaluation techniques for LLM pipelines (e.g., hallucination detection, prompt testing frameworks).
