Role Description The Perception team builds the system which learns the spatial-temporal representation and their semantic meanings of the surrounding environment of the autonomously driving vehicle (ADV). We work jointly with downstream teams on the optimization and integration into the Waymo Driver. We conduct our own research to address real-world problems and collaborate with research teams at Alphabet. - Design, implement, and optimize large-scale continual pre-training pipelines for cutting-edge VLM foundation models. - Conduct research and development on novel pre-training techniques, focusing on efficiently integrating new, diverse, and multimodal data streams (e.g., visual data from different sensors) into existing models. - Develop and rigorously evaluate metrics and methodologies for measuring the performance, and transferability of continually pre-trained foundation models in the context of autonomous driving. - Stay current with the latest advancements in large language models, vision-language models, and continual learning, and translate relevant research into production-ready systems. Qualifications - 5+ years of experience in Machine Learning, with a focus on large-scale model development (LLM, VLM, or similar foundation models). - Proven expertise in LLM/VLM pre-training, continual learning with large scale datasets. - Strong coding proficiency in Python and deep learning frameworks (e.g., Jax, TensorFlow, PyTorch). - Hands-on experience with model training, evaluation, and deployment in a production environment. - Master's degree in Computer Science, Electrical Engineering, or a related field, or equivalent practical experience. Requirements - Experience in fine-tuning foundation models for autonomous driving or robotics applications. - Familiarity with large-scale data curation and quality assurance processes for multimodal datasets. - Background in autonomous vehicle perception, motion planning, or decision-making systems. - Publications in top-tier machine learning or computer vision conferences (e.g., NeurIPS, ICML, CVPR, ICCV, ECCV). - PhD in a relevant field. Benefits - Waymo employees are eligible to participate in Waymo’s discretionary annual bonus program. - Equity incentive plan. - Generous Company benefits program, subject to eligibility requirements. Salary Range The expected base salary range for this full-time position across US locations is $170,000 — $216,000 USD. Actual starting pay will be based on job-related factors, including exact work location, experience, relevant training and education, and skill level.

• Work as a product owner to deliver on the product vision and feature priorities. You should have a strong track record of making tough tradeoffs to balance scope, quality, supportability, performance, and time criticality. • Guide the team through design/implementation for complex technical projects. • Work closely with the internal stakeholders to ensure the product meets quality/stability requirements of enterprise customers, leveraging experience inventing and improving technology of performance/stability/scale. • Manage end to end product ownership, from planning and design to on call product support. Manage/fix/communicate issues that arise during escalations/customer issues.

• Participate in data discovery workshops to inventory source systems including property management platforms, marketing channels, and CRM data, and translate findings into data lake architecture requirements. • Design and implement a multi-zone enterprise data lake on Amazon S3 (raw, conformed, enriched, aggregated) with ingest, cleansing, and business layers aligned to the SOW architecture. • Build batch and streaming data ingestion pipelines using AWS Glue, Amazon Kinesis, and AWS Data Pipeline across CDP, marketing, and property management data sources. • Implement data transformation and orchestration frameworks using AWS Glue ETL and AWS Step Functions, including AWS Glue Data Catalog for metadata management and discovery. • Configure Amazon Athena for serverless SQL querying across the data lake; support QuickSight integration with curated data sets for business analytics. • Develop and deploy ML models on Amazon SageMaker for lead scoring, predictive maintenance, intelligent underwriting risk scoring, and AI-powered audience segmentation. • Integrate Amazon Bedrock foundation models to enable generative AI capabilities including customer profile enrichment, hyper-personalization, and intelligent marketing automation. • Use Kiro CLI to accelerate AI-assisted development workflows, spec-driven pipeline implementation, and automated code generation tasks. • Design and implement entity resolution pipelines using Amazon Entity Resolution to identify, deduplicate, and merge customer records into unified golden records. • Implement real-time and batch data synchronization pipelines between source systems and the Customer Data Platform (CDP). • Support Azure data lake migration: conduct discovery, assess schemas and transformation logic, provision AWS target environments, execute migration via AWS DataSync, and perform data validation and reconciliation. • Implement data lake security using AWS Lake Formation, including row-level security and column-level encryption. • Build and maintain data models to support Customer 360 views, ML feature stores, and executive analytics dashboards. • Ensure data quality, validation, and integrity across all pipeline stages and ML model outputs; support UAT for data-dependent features. • Collaborate with Full Stack, DevOps/MLOps, and AWS engagement teams; contribute to architecture documentation, pipeline runbooks, and data governance documentation.

Role Description Senior AI/ML Engineers are senior individual contributors who design, build, and deploy production-grade AI/ML systems for both client-facing and internal products. They partner with leadership and cross-functional teams to translate business needs into scalable ML and LLM-based solutions. This role does not typically include direct reports but requires strong technical leadership, mentorship, and influence across teams. Responsibilities - AI/ML System Design & Leadership - Lead the design and implementation of scalable ML systems, including supervised, unsupervised, and LLM-based solutions. - Translate research and prototypes into production-ready systems. - Partner with stakeholders to identify high-impact AI/ML opportunities and define optimal technical approaches. - Provide technical mentorship and contribute to team upskilling. - LLM & Production AI Systems - Build and operate LLM pipelines, including prompt design, fine-tuning, and evaluation. - Develop RAG-based systems using embeddings, vector stores, and retrieval strategies. - Design evaluation frameworks, feedback loops, and datasets to continuously improve model performance. - Create reusable tooling to accelerate experimentation, deployment, and monitoring. - MLOps & Deployment - Own end-to-end ML lifecycle: data pipelines, training, deployment, monitoring, and iteration. - Establish best practices for reproducibility, observability, CI/CD, and model versioning. - Partner with platform/DevOps teams to ensure reliability and scalability. - Promote responsible AI practices, including governance, fairness, and transparency. - Cross-Functional Collaboration - Lead cross-functional initiatives across data engineering, analytics, and AI/ML. - Translate complex ML concepts into clear recommendations for technical and non-technical audiences. - Collaborate with clients and internal teams to plan and deliver AI/ML solutions. - Contribute documentation, frameworks, and shared best practices. - Project Execution - Scope and lead complex AI/ML initiatives aligned to business outcomes. - Align stakeholders and drive execution across teams. - Establish clear success metrics and ensure delivery of high-impact solutions. Qualifications - 5–7 years of experience in ML engineering, AI engineering, or related fields, with production deployment experience. - Strong programming skills in Python and SQL; experience with PyTorch and HuggingFace. - Experience building LLM applications, including RAG, embeddings, and vector search. - Experience with cloud platforms (AWS or Azure; e.g., SageMaker, Bedrock, Azure ML). - Strong understanding of ML fundamentals: data design, training, evaluation, and experimentation. - Familiarity with LLM alignment techniques (e.g., SFT, DPO, RL). - Experience with MLOps practices: CI/CD, monitoring, retraining, and experiment tracking. - Proficiency working with complex, multi-source datasets and defining evaluation strategies. - Strong software engineering fundamentals (testing, modularity, code review). - Experience mentoring engineers and influencing technical direction. - Strong communication skills with both technical and non-technical stakeholders. Tech Stack - Languages: Python, SQL - Frameworks: PyTorch, HuggingFace - Platforms: AWS, Azure, Snowflake, Databricks Physical Requirements - Frequent sitting at a desk performing work on a computer. - Reasonable accommodations may be made to enable individuals with disabilities to perform the essential functions. Compensation Compensation Range: $175,000 - 190,000 annually. Please note that compensation packages are finalized after the interview process is concluded. We use a competency-based approach to base pay, which means it is based on the competencies and skills demonstrated for this role. Core Company Values - Take the Long View - Ensure the company is built to last. - Be Courageous - Make the right decisions even when they aren't the easiest decisions. - Be Genuine - Bring honesty and authenticity to all that you do. - Work with Focus + Passion - Display purpose and pride in your work and never stop learning. As an equal opportunity employer, we are firmly committing to diversity, equity, and inclusion in our hiring efforts. We recognize that we need team members from all backgrounds and experiences to successfully shape a positive employee experience as well as deliver our product and service solutions. To that end, we actively seek candidates who can bring diverse experiences and backgrounds to our team. We know that complex factors and systemic bias can get in the way of us meeting strong candidates, so please don't hesitate to apply even if you're not 100% sure. At this time, Velir does not sponsor candidates and unfortunately cannot accept those on OPT or CPT.