AI Product Engineer (LLM Integration – SaaS)

About the Role

We’re looking for a hands-on AI Product Engineer with experience working with Large Language Models (LLMs) to help us bring AI capabilities into our B2B SaaS platform. This role sits at the intersection of product engineering, machine learning, and user experience — helping deliver meaningful, adoptable AI features to our customers.

You’ll work closely with product managers, designers, and backend engineers to define how LLM-powered features (e.g., summaries, suggestions, chat agents) are designed, built, and adopted by real users.

What You’ll Do

• Design and develop product features powered by LLMs (e.g., OpenAI, Bedrock, Mistral, Claude)
• Create pipelines for prompt engineering, input pre-processing, and response parsing
• Integrate LLMs via APIs and manage token usage, latency, and fallback behavior
• Collaborate with product and UX to scope AI-driven capabilities that improve usability and value
• Track performance of LLM outputs (relevance, accuracy, completion rate) and refine prompts or logic accordingly
• Contribute to customer adoption efforts — building explainable, reliable, and intuitive AI features

What We’re Looking For

• Solid experience with LLM APIs (OpenAI, Anthropic, AWS Bedrock, etc.)
• Strong coding skills in Python, Node.js, or TypeScript
• Experience working with SaaS applications — especially B2B use cases
• Understanding of prompt engineering, embeddings, RAG, and vector search (e.g., Pinecone, Weaviate, FAISS)
• Familiarity with UI/UX principles in AI adoption (what makes features feel useful, not gimmicky)
• Ability to work independently in a remote, fast-paced environment

Nice to Have

• Experience with LangChain, LlamaIndex, or similar orchestration frameworks
• Knowledge of data privacy and AI governance in production
• Background in ML/NLP pipelines beyond API usage (e.g., fine-tuning, embeddings, model evaluation)

Why Join Us?

• Work remotely with a global, product-driven SaaS team
• Help shape how AI is adopted meaningfully, not just bolted on
• Build capabilities that customers use and trust — not experiments
• Flexible schedule, strong ownership, and room to explore new ideas