DataJobs.io
← Back to all jobs

Job Description

This Senior Data Scientist / ML Engineer role at Robotics Technologies focuses on transforming traveler and catalog data into narrative text and synthetic data, then building scalable deep learning and LLM fine-tuning and inference pipelines on AWS. The position is based in Sanjose, CA (onsite) and emphasizes distributed training, rigorous evaluation, and production-focused serving.

Responsibilities

  • Convert unstructured and structured traveler and catalog data into narrative text documents and synthetic data, operating at a scale of tens of millions of documents.
  • Own feature and context engineering decisions, including what information each document contains and how it is expressed (for example, market-relative price and tier and trip context).
  • Lead large-scale distributed training using Amazon SageMaker AI, including HyperPod.
  • Drive LLM fine-tuning workflows using LoRA/QLoRA adapters, supervised training, and reinforcement learning approaches for ranking and reasoning.
  • Build and iterate on deep learning models, including architecture design, ranking heads, co-training, and multi-task setups.
  • Develop low-latency, cost-efficient inference and deployment strategies, including quantization, distillation, and profiling.
  • Perform evaluation and error analysis by building evaluation sets, running ablations, and identifying data leakage; document findings for the customer and the AWS science team.
  • Handle data privacy practices on AWS, including removing per (privacy-related personal data removal is required).

Requirements

  • Experience with Spark (PySpark) and the Hadoop ecosystem, including Hive and HDFS, working on corpora containing tens of millions of documents.
  • Hands-on Amazon SageMaker AI experience, including processing, training, distributed training with HyperPod, and deployment.
  • LLM fine-tuning experience using LoRA/QLoRA, plus reinforcement learning methods such as GRPO/PPO-style approaches.
  • Strong deep learning foundations in PyTorch, including model architectures, training dynamics, and co-training.
  • Experience with large-scale batch LLM inference and production inference using vLLM and/or Amazon Bedrock batch.
  • Working knowledge of data privacy and security on AWS, including personal-data removal, IAM, KMS, and Lake Formation.
  • Ability to communicate clearly in writing with customer and science stakeholders.

Technologies

  • Amazon SageMaker AI, HyperPod
  • LoRA, QLoRA
  • PyTorch
  • Spark, PySpark, Hadoop ecosystem, Hive, HDFS
  • vLLM, Amazon Bedrock, Amazon Bedrock batch
  • IAM, KMS, Lake Formation
  • Amazon Trainium, Inferentia, Neuron SDK
  • GRPO, PPO
  • Quantization, distillation, profiling
  • Reinforcement learning

Preferred Skills

  • Experience with AWS Trainium and Inferentia via the Neuron SDK.
  • Knowledge of recommender systems, including ranking metrics such as NDCG and MRR, along with sequential and generative recommenders.
  • Experience with embedding quantization or semantic IDs.
  • Experience in a customer-facing or embedded role.

Similar Jobs