Software Development Engineer II, Generative AI

Verisk Analytics • Jersey City, NJ

4d • Hybrid

About The Position

This is a hands-on engineering role focused on building production-grade generative AI systems rather than prototypes. The engineer will contribute across the full lifecycle, from experimentation and prompt and RAG design through production deployment, observability, and iteration, working closely with senior engineers, product owners, and technical leadership. At the SDE II level, the role takes ownership of medium-complexity features end-to-end, mentors SDE I engineers, and contributes to architecture and design decisions for GenAI components.

Requirements

• At least 3 years of industry experience in software engineering, data science, or machine learning.
• At least 1 year of hands-on experience building GenAI/LLM-based features or applications, ideally in a production setting.
• Strong Python skills and experience with modern AI/ML and GenAI frameworks and tooling.
• Working understanding of LLMs, embeddings, RAG, prompt engineering, context engineering, and GenAI architecture patterns.
• Experience with cloud-native architectures and deploying scalable applications to production.
• Understanding of GenAI evaluation, testing, monitoring, and responsible AI practices.
• Ability to work with moderate autonomy, break down ambiguous problems, and communicate technical tradeoffs clearly.
• Curiosity and enthusiasm for learning new technologies, tools, and techniques in a fast-moving field.
• Bachelor's degree in Computer Science, Software Engineering, Machine Learning, Data Science, or a related technical field.

Nice To Haves

• Knowledge of fine-tuning or adapting LLMs (e.g., SFT, LoRA/PEFT, instruction tuning), including preparing training datasets or evaluating fine-tuned models against baselines.
• Experience with agentic AI patterns or multi-step LLM workflows.

Responsibilities

• Build features and components for production-grade GenAI products, with a focus on quality, reliability, and performance.
• Implement and iterate on LLM-based solutions, including retrieval-augmented generation (RAG), prompt engineering, context engineering, loop engineering, and agentic workflows.
• Write clean, well-tested, maintainable Python code and contribute to the broader codebase and infrastructure.
• Help build evaluation and monitoring pipelines to measure output quality, groundedness, and reliability of GenAI features.
• Work with foundation models and GenAI frameworks and tooling, evaluating tradeoffs between approaches under guidance from senior engineers.
• Collaborate with Product Owners and cross-functional stakeholders to translate requirements into working technical solutions.
• Participate in code reviews, design discussions, and engineering best practices across the GenAI development lifecycle.
• Debug and resolve technical issues across the stack, escalating and collaborating with senior team members on complex problems.
• Take ownership of medium-complexity features end-to-end, mentor SDE I engineers, and contribute to architecture and design decisions for GenAI components.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level