Applied AI Engineer

We are looking for a dynamic Software Engineer to join our AI Agents team. Bicycle leverages AI Agents in pivotal roles, functioning as Business Analysts and Data Analysts who collaborate to uncover data-driven insights and recommend actionable strategies to our customers. In this role, you will be responsible for building agents that efficiently work with vast volumes of data across various industries. These agents are designed to automate numerous data analysis tasks, including data transformation and analysis, thereby enhancing the efficiency of data-driven operations. The ideal candidate will have prior experience with data analysis and a passion for using innovative technologies such as LLM and AutoML to craft creative solutions that automate the most repetitive and time-consuming parts of data analysis.

Responsibilities

• Design, develop, and implement AI agents that act as business and data analysts within our platform.
• Collaborate with cross-functional teams to enhance agent capabilities and ensure seamless data-driven insights delivery.
• Utilize Agentic RAG to decompose complex queries and strategically improve information retrieval processes.
• Implement and manage multi-agent frameworks to support autonomous task execution.
• Integrate external APIs and enhance agent memory systems for improved contextual awareness.
• Evaluate and optimize multi-agent workflows to improve performance and efficiency.
  
  

Requirements

• BS/MS degree in Computer Science, AI, or a related field.
• 2-3 years of production-grade Python development experience.
• Hands-on experience with data analysis using Python.
• Proficiency in working with Large Language Models (LLMs) and an understanding of prompt engineering techniques.
• Experience with Retrieval Augmented Generation (RAG) systems and Agentic RAG approaches.
  
  

Good To Have

• Familiarity with frameworks such as Langgraph, Crew AI, and vector databases.
• Knowledge of specialized task agents focused on domain-specific problems.
• Exposure to dynamic tool selection and parallel processing in multi-agent environments.
• Knowledge of API integration and memory management for contextual decision-making.
• Familiarity with workflow design for multi-agent systems and evaluation metrics for assessing agent performance.
  
  

This job was posted by Venkatesh Ponnuswamy from Bicycle AI.