AI Engineer - United States

• LLM Integration & Development: Design, develop, and deploy production-grade applications leveraging various LLMs, context optimization, etc.
• Agentic Workflows: Architect and implement sophisticated, multi-step and multi-agent workflows using frameworks like LangChain and LangGraph.
• Retrieval Augmented Generation (RAG): Build and optimize RAG pipelines, including implementing and managing embeddings, vector databases, and advanced rerankers to enhance response quality and relevance.
• Full-Stack Development: Own end-to-end implementation from frontend to the backend.

This role helps purpose-driven enterprises manage and improve their environmental, social, and governance (ESG) and overall sustainability footprint through innovative AI technology, contributing to positive planetary impact.

• 5+ years of professional experience in software development, with a minimum of 2 years focused on AI/ML development, particularly with LLMs
• Strong proficiency in Python and its relevant data science libraries (e.g., Pandas, NumPy, Scikit-learn)
• Proven experience integrating and working with major LLM APIs - both public and private/local, e.g., Gemini, OpenAI, Anthropic, Llama, Ollama, etc., including hands-on experience using techniques for LLM efficiency
• At least 1 year of deep practical experience with LangChain and LangGraph for building complex LLM applications and agentic workflows using autonomous agents, tools, memory management, parallelization, etc.

• Salary, equity, and benefits
• Hybrid-friendly work environment, in office 2-3 times per week (for those in New York, San Francisco Bay Area, and Munich)
• High ownership and autonomy
• Fast-moving, execution-focused culture
• Work at the cutting edge of AI-driven software development

Pulsora is an Enterprise Sustainability Platform startup dedicated to empowering purpose-driven enterprises to manage and improve their environmental, social, and governance (ESG) and overall sustainability footprint with an integrated technology platform.