At Allstate, great things happen when our people work together to protect families and their belongings from life’s uncertainties. And for more than 90 years, our innovative drive has kept us a step ahead of our customers’ evolving needs. From advocating for seat belts, air bags and graduated driving laws, to being an industry leader in pricing sophistication, telematics, and, more recently, device and identity protection.
Job Description
In this role, you will design and build AI-driven Insight Studio for the Scholar product, transforming large-scale customer conversations into actionable, real-time insights. You will hands-on build and productionize AI systems, integrating LLMs, RAG pipelines, and AI agents into scalable, reliable applications that directly power business decisions.
Key Responsibilities
Translate complex business problems into end-to-end AI solutions, leveraging machine learning and Generative AI techniques aligned with product and business goals.
Design, build, and operate production-grade AI systems for conversational understanding, summarization, classification, and insight generation from large-scale customer interaction data.
Architect and implement Retrieval-Augmented Generation (RAG) pipelines, integrating structured and unstructured data sources to deliver context-aware, grounded, and explainable AI outputs.
Build and optimize vector search and embeddings-based systems, enabling semantic search, similarity matching, and efficient information retrieval at scale.
Design and implement AI agent workflows, orchestrating multi-step reasoning, tool usage, and decision-making using frameworks such as LangChain, LangGraph, or similar.
Apply prompt engineering and LLM orchestration techniques to improve output quality, consistency, latency, and cost efficiency in production environments.
Integrate foundation models via platforms such as AWS Bedrock, including fine-tuning, evaluation, and deployment strategies tailored to business use cases.
Build and maintain scalable data pipelines and distributed processing systems to support real-time and batch AI workloads.
Own end-to-end lifecycle of AI/ML solutions, including data preparation, model development, evaluation (including LLM evaluation and guardrails), deployment, and continuous improvement.
Implement MLOps and LLMOps practices, including CI/CD pipelines, versioning, monitoring, and automated evaluation for both traditional ML and AI systems.
Monitor and optimize AI systems using observability tools (e.g., Datadog, CloudWatch), ensuring performance, reliability, and cost control in production.
Analyze model outputs, communicate findings, and iterate on solutions to continuously improve business impact and user experience.
Essential Skills
Data Science & AI Expertise
Strong proficiency in Python for AI/ML
Proven experience building and scaling production-grade AI/ML systems
Deep understanding of data pipelines, distributed systems, and scalable architectures
Strong grounding in MLOps / LLMOps workflows and lifecycle management
Practical experience building applications using: Large Language Models (LLMs) (prompting, fine-tuning, orchestration)
Retrieval-Augmented Generation (RAG) architectures
Embeddings and vector search systems
AI agents and multi-step reasoning workflows
Strong understanding of LLM evaluation, guardrails, and responsible AI practices
Frameworks & Tooling
Hands-on experience with: LangChain, LangGraph, LlamaIndex or similar orchestration frameworks
Experience designing modular, reusable AI pipelines and components
Data & Storage
Experience with vector databases and modern data storage systems
Familiarity with handling large-scale structured and unstructured datasets
Cloud & AI Platforms
Experience with: Amazon SageMaker or similar ML platforms
AWS Bedrock for foundation model access and deployment
AWS services such as Lambda, S3, ElastiCache
Experience deploying LLM-powered applications in cloud-native architectures
Observability &…