AI Engineer (Agentic Systems)
Role details
Job location
Tech stack
Job description
- Architect and develop LangChain/LangGraph-based multi-agent systems for enterprise workloads.
- Design and expose backend services and orchestration APIs using FastAPI and Python.
- Engineer scalable agentic workflows capable of processing and reasoning over large volumes of data.
- Build and optimize Retrieval-Augmented Generation (RAG) workflows integrating vector databases and relational data stores.
- Apply expert-level prompt and context engineering techniques to maximize LLM reliability and performance.
- Implement observability, telemetry, and monitoring instrumentation to ensure the operational health of agentic services.
- Maintain high code quality standards using Python best practices and Pydantic for data validation.
- Proactively identify technical debt, architectural gaps, and new workstreams, owning them through to completion.
Requirements
Experience: 5+ years of software engineering experience with a strong focus on AI/ML engineering and backend systems. Deep hands-on experience with LangChain and LangGraph for building agentic and multi-agent systems is required.
Technical Skills: Expertise in Python and Pydantic is necessary. A strong understanding of Large Language Models (LLMs), including architecture, behavior, limitations, and prompt engineering, is essential. Experience integrating vector databases into RAG workflows, particularly Elasticsearch, is strongly preferred. Candidates must have demonstrated ability to design agentic workflows that handle big data processing and experience building and consuming RESTful APIs with FastAPI.
Other Qualifications: Must be self-directed and autonomous, with the ability to own a service end-to-end and execute without heavy oversight. A conceptual understanding of LLM fine-tuning and training concepts is required.
Preferred Qualifications
- Frontend experience with React, HTML, and/or TypeScript.
- A background in Data Science or familiarity with data pipelines and analytics.
- Experience with LangSmith for LLM tracing, debugging, and evaluation.
- Experience with Galileo for LLM observability and monitoring.
- Experience working in enterprise or financial services engineering environments.