Senior C++ Software Engineer with CUDA/GPU/TPU

EPAM Systems, Inc.
Newtown, PA, United States
12 days ago

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Experience level
Expert
Experience required
5 years minimum
Working hours
Regular working hours
Languages
English
Job source

Tech stack

Artificial Intelligence Artificial Neural Networks C++ (Programming Language) Nvidia CUDA Computer Programming Machine Learning Tensorflow Software Engineering Application Specific Integrated Circuits Pytorch Large Language Models Deep Learning
+1 more
Unsupervised Learning

Job description

  • Write kernels in C++ for various operations used across AI and ML workloads
  • Develop and optimize the compiler and neural network stack supporting LLMs and CNNs
  • Collaborate with experts in computer architecture, ASIC design and advanced systems to build next-generation computing solutions
  • Design and implement AI/ML kernel operations tailored to GPU/AI-accelerator architectures
  • Build and maintain ML models using PyTorch or TensorFlow
  • Ensure kernel-level performance and correctness across modern neural network architectures such as transformers
  • Take ownership of key technical decisions as a lead developer within the ML & AI team

Requirements

  • 5+ years of experience in software engineering with a focus on ML/AI
  • 5+ years of experience in C++ and CUDA/kernel programming
  • Knowledge in GPU/AI-accelerator architectures
  • Experience in building and working with ML models in PyTorch or TensorFlow
  • Understanding of modern ML model architectures such as transformers
  • English proficiency at B2 level or higher

Nice to have

  • Proficiency in Python programming with hands-on experience in developing, training and fine-tuning deep learning models
  • Skills in designing, implementing and iterating on neural network architectures to achieve optimal performance on diverse tasks
  • Capability to investigate and troubleshoot model performance issues including gaps in compilation steps and kernels
  • Demonstrated experience in designing, training and deploying neural networks for various applications
  • Solid understanding of machine learning fundamentals including supervised and unsupervised learning techniques

Benefits & conditions

  • We gather like-minded people: *

  • Top tech minds driving innovation in AI, cloud and digital platform modernization
  • Supportive team and agile, startup-like culture
  • Hybrid by design mode and opportunity to work remotely within Poland
  • Chance to work abroad for up to 60 days annually
  • Business-driven relocation opportunities
  • We provide growth opportunities: *

  • Career development programs
  • Thought leadership, mentoring, soft skills and well-being programs
  • Certification (Anthropic, Gemini, GCP, Azure, AWS)
  • English classes
  • We cover it all: *

  • Stable pay
  • Participation in the Employee Stock Purchase Plan with a 15% discount
  • Benefits package (health insurance, multisport, shopping vouchers)
  • Referral bonuses up to $2,000
  • Offices featuring entertainment and relaxation zones, table tennis and football, free snacks, coffee and more
  • Corporate, social and well-being events
  • Please, note: *

  • Benefits listed above are available to employees only
  • We are open for working with Contractors. Terms of B2B cooperation agreements are agreed individually
  • We will reach out to selected candidates exclusively

About the company

EPAM is global leader in AI transformation engineering and integrated consulting, serving Forbes Global 2000 companies and ambitious startups. With over thirty years of expertise in custom software, product and platform engineering, we empower our clients to become AI-Native enterprises, driving measurable value from innovation and digital investments.

About EPAM Systems

501-1000

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on arc.dev

Good distractions

Talks and stories from around this role — technically off-topic, practically not.

2:33 min

Architecting CUDA and the AI software stack

Michael Kagan Michael Kagan +1 · WWC Europe 2026

2:35 min

Preventing remote code execution in PyTorch models

Balázs Kiss · WWC 2023

1:39 min

Fundamentals of tensors and the TensorFlow library

Håkan Silfvernagel · LIVE

6:21 min

Previewing upcoming hardware acceleration capabilities for Python environments

Chris Heilmann +2 · LIVE

3:30 min

Transitioning from CUDA software architect to user

Stephen Jones · Coffee With Developers

1:37 min

Accelerating compute with focused developer tools

Julia Koch Julia Koch +1 · WWC Europe 2026

Videos

See all

Related articles

See all