Senior AI/ML Engineer - Research

84.51 Overview:

84.51 is a retail data science, insights and media company. We help The Kroger Co., consumer packaged goods companies, agencies, publishers and affiliates create more personalized and valuable experiences for shoppers across the path to purchase.

Powered by cutting-edge science, we utilize first-party retail data from more than 62 million U.S. households sourced through the Kroger Plus loyalty card program to fuel a more customer-centric journey using 84.51 Insights, 84.51 Loyalty Marketing and our retail media advertising solution, Kroger Precision Marketing.

84.51 follows a 5-day in-office work schedule to support collaboration, alignment, and team connection.

Join us at 84.51 !

Senior AI/ML Engineer P4368)

We are seeking a highly skilled Senior AI/ML Engineer to join our AI Foundation Models team, focused on pushing the boundaries of AI capabilities through advanced reasoning, multi-step problem solving, and deep agentic systems. In this role, you will drive experimentation and rapid iteration across LLMs, deep learning, reinforcement learning, world modeling, and causal reasoning to solve complex retail problems at scale. You will own the full lifecycle of these systems-from prototyping and experimentation through production-grade deployment, optimization, and monitoring.

Responsibilities

Design, develop, and deploy end-to-end deep reasoning and research agents capable of complex, multi-step problem solving for the retail domain

Architect agent systems leveraging test-time compute scaling strategies to enhance reasoning quality during inference

Develop and apply reinforcement learning techniques to improve agent reasoning capabilities and alignment; design reward functions, preference models, and human feedback pipelines for complex reasoning tasks

Research and experiment with world modeling and causal reasoning approaches to enable agents to understand cause-effect relationships and simulate outcomes

Lead research in model pretraining, fine-tuning, instruction tuning, and parameter-efficient methods (LoRA, adapters, QLoRA); implement novel architectures and prompting strategies across LLMs and SLMs

Build and optimize encoder-only architectures, embedding models, and dense retrieval systems for downstream agent capabilities

Write production-quality, scalable code for training, inference, and deployment; implement distributed training systems optimized for efficiency, memory, and latency

Develop evaluation frameworks and benchmarking systems for reasoning quality, agent reliability, and task completion

Collaborate with cross-functional teams including researchers, product teams, and infrastructure; mentor junior engineers

Qualifications

Education & Experience

Masters in Computer Science, Machine Learning, Artificial Intelligence, or related field

1-2 years of industry or research experience in deep learning

Required Skills

Hands-on experience building agentic systems, multi-step reasoning systems, or research agents using LLMs, with strong understanding of agentic design patterns including planning, tool use, memory, RAG, and self-reflection

Proven experience with RLHF, RLEF, reward modeling, PPO, DPO, or related alignment and RL techniques

Familiarity with world modeling, causal inference, and causal reasoning frameworks applied to decision-making and planning

Expert proficiency in PyTorch with deep experience in transformer architectures and modern LLM/SLM implementations; familiarity with distributed training frameworks (DeepSpeed, FairScale, Megatron)

Experience in pretraining large models including data preprocessing, tokenization, training dynamics, fine-tuning, and parameter-efficient methods

Knowledge of encoder-only architectures, masked language modeling, embedding models, and dense retrieval

Clean, efficient, scalable Python coding skills with knowledge of model quantization, pruning, distillation, and compression techniques

Experience with experiment tracking (MLflow), deployment pipelines, cloud platforms (GCP, Azure), and containerization (Docker, Kubernetes)

Strong problem-solving skills, ability to work independently on ambiguous problems, and excellent communication skills

Preferred Skills

Experience with multimodal models, cross-modal reasoning, and vision-language agents

Familiarity with agent orchestration tools (LangChain, LangGraph, AutoGen, or custom frameworks)

Contributions to open-source deep learning or agentic AI projects

Experience with hardware optimization (GPUs, TPUs), mixed-precision training, and synthetic data generation

Background in NLP applications, information retrieval, or knowledge-intensive tasks

#LI-SSS

Pay Transparency and Benefits

The stated salary range represents the entire span applicable across all geographic markets from lowest to highest. Actual salary offers will be determined by multiple factors including but not limited to geographic location, relevant experience, knowledge, skills, other job-related qualifications, and alignment with market data and cost of labor. In addition to salary, this position is also eligible for variable compensation.

Below is a list of some of the benefits we offer our associates:

Health: Medical: with competitive plan designs and support for self-care, wellness and mental health. Dental: with in-network and out-of-network benefit. Vision: with in-network and out-of-network benefit.

Wealth: 401(k) with Roth option and matching contribution. Health Savings Account with matching contribution (requires participation in qualifying medical plan). AD&D and supplemental insurance options to help ensure additional protection for you.

Happiness: Paid time off with flexibility to meet your life needs, including 5 weeks of vacation time, 7 health and wellness days, 3 floating holidays, as well as 6 company-paid holidays per year. Paid leave for maternity, paternity and family care instances.

Pay Range

$98,000-$169,050 USD