Software Engineer - Machine Learning Platform

Salary: $61,000 - 101,000 per year Requirements:

• We require formal training or certification in software engineering concepts, along with 3+ years of applied experience.
• We want demonstrated experience running ML training in cloud environments and troubleshooting issues across infrastructure and code.
• We need strong Python skills and solid engineering discipline, including testing, code reviews, modular design, and dependency management.
• We look for experience building automation and CI pipelines for ML codebases, including build, test, release, and deployment or promotion workflows.
• We require hands-on experience with deep learning training workflows and at least one major framework such as PyTorch or TensorFlow.
• We expect a strong understanding of training performance and stability, including data loading bottlenecks, mixed precision, checkpointing, reproducibility, and evaluation methods.
• We require experience with distributed training concepts such as DDP, FSDP, or DeepSpeed, plus familiarity with collective communication, scaling, and bottleneck analysis.
• We need the ability to profile and optimize training systems across CPU and GPU utilization, memory, I/O throughput, networking, and scheduling.
• We require experience with Kubernetes fundamentals for compute-intensive workloads and AWS services such as EKS, ECR, S3, IAM, VPC/networking, CloudWatch, and EC2.
• We require hands-on experience using enterprise-authorized AI-assisted development tools, with the ability to critically review, validate, and refine AI-generated output for correctness, performance, and security.
• We expect understanding of responsible AI practices in engineering workflows, including data sensitivity, secure handling of inputs and outputs, and resiliency and security expectations, plus the ability to guide peers on safe usage.
• Preferred: experience running training workloads across multiple cloud platforms and managing portability, performance, and governance between environments.
• Preferred: familiarity with cloud-native networking and storage patterns for high-throughput training and artifact management.
• Preferred: experience optimizing training input pipelines, including sharding, prefetching, caching, and formats such as Parquet or WebDataset, as well as working with large datasets.
• Preferred: familiarity with distributed compute frameworks such as Spark, Ray, or Dask for feature and dataset generation.
• Preferred: familiarity with workflow orchestration tools and model registry concepts.
• Preferred: experience improving training cost and performance through right-sizing, scheduling policies, interruptible capacity strategies where applicable, budget guardrails, and quota planning.
• Preferred: strong observability practices for training systems, including metrics, logs, traces, GPU telemetry, dashboards, and alert tuning.

• We will design, build, and maintain an end-to-end ML training platform.
• We will run and tune GPU training workloads, both single-node and distributed, to improve throughput, utilization, and reproducibility.
• We will build and operate training infrastructure on Kubernetes platforms such as EKS, including resource management and workload troubleshooting.
• We will enable Gen AI and LLM training and fine-tuning workflows, including supervised fine-tuning, evaluation harnesses, artifact and version governance, and scalable GPU execution aligned with enterprise controls.
• We will implement observability for training systems through metrics, logs, dashboards, alerting, and operational runbooks.
• We will partner with data engineering and platform teams to define interfaces, standards, and guardrails for security, access, and cost control.
• We will improve the developer experience for training through standardized containers, CI/CD, templates, documentation, and self-service workflows.
• We will use enterprise-authorized AI coding assist tools to improve code quality, delivery speed, and productivity, while validating outputs through peer review, automated testing, and secure coding standards.
• We will apply knowledge of Software Development Life Cycle tooling, including enterprise-authorized AI-assisted development and automation capabilities, to increase the value delivered by automation.

• AI
• AWS
• CI/CD
• Cloud
• CloudWatch
• EC2
• Fine-tuning
• IAM
• Support
• Kubernetes
• LLM
• Marketing
• PyTorch
• Python
• Security
• Spark
• TensorFlow
• NodeJS
• Airflow

More:

hackajob is collaborating with J.P. Morgan to connect exceptional professionals with this opportunity. We are offering a Software Engineer III position within our AI/ML data platform team, where we build and operate scalable, reliable ML training systems and pipelines on AWS and other cloud platforms. In this role, we focus on productionizing training workloads, improving performance and cost efficiency, and enabling repeatable, well-governed training across environments. We offer a competitive total rewards package, including base salary determined by role, experience, skill set, and location, with eligible roles also receiving commission-based pay or discretionary incentive compensation. Our benefits and programs may include comprehensive health care coverage, on-site health and wellness centers, a retirement savings plan, backup childcare, tuition reimbursement, mental health support, financial coaching, and more. We value diversity and inclusion, provide equal opportunity employment, and support reasonable accommodations. This role is also subject to Section 19 of the Federal Deposit Insurance Act, including review of criminal conviction history as part of the hiring process. Our Corporate Functions team supports finance, risk, human resources, marketing, and other essential business areas that help set our businesses, clients, customers, and employees up for success.

last updated 41 week of 2026