Description:
We are seeking a AI/ML Engineer (Agentic AI) to architect and implement autonomous, goal-driven AI agents. This role is ideal for someone experienced in multi-agent systems, orchestration frameworks, and deploying complex LLM-based agents that can reason, plan, and act in dynamic environments.
Key Responsibilities
- Agentic System Design – Design and deploy multi-agent or hierarchical reasoning systems capable of planning, self-correction, and collaboration.
- LLM Orchestration – Build long-running pipelines using frameworks such as LangGraph, AutoGen, or similar to manage memory, tool-use, and context.
- Tool & Data Integration – Connect agents to APIs, knowledge graphs, and vector databases (e.g., Pinecone, Weaviate, Milvus) to enable autonomous decision-making.
- Data & Infrastructure – Develop and maintain data pipelines and embedding stores; implement ETL/ELT workflows.
- Optimization & Scaling – Fine-tune and optimize LLMs for performance, cost, and safety; implement monitoring and feedback loops.
- Cross-Functional Collaboration – Partner with product, engineering, and compliance teams to align agent behavior with business goals and regulatory requirements.
Must-Have Qualifications
- Demonstrated experience designing and deploying agentic or multi-agent AI systems.
- Deep understanding of transformer architectures and advanced prompt-engineering strategies.
- Strong Python skills and expertise with orchestration frameworks (LangGraph, AutoGen, CrewAI, etc.).
- Proficiency in building robust data pipelines and working with vector databases.
- Cloud deployment and MLOps experience (AWS, GCP, or Azure).
Preferred Qualifications
- Experience fine-tuning LLMs with LoRA/QLoRA/PEFT.
- Knowledge of reinforcement learning, planning algorithms, or neuro-symbolic methods.
- Contributions to open-source agentic AI frameworks or published research in autonomous AI.