AI Performance Optimization Engineer
Not Disclosed•Full-TimeRemote
location_on3155;3158, Cardenas Terrace, Centerville District, Fremont, Alameda County, California, 94536, United States
About the Team
Bright Vision Technologies is a forward-thinking software development company dedicated to building innovative solutions that help businesses automate and optimize their operations. We leverage cutting-edge technologies to create scalable, secure, and user-friendly applications. As we continue to grow, our team remains dynamic and mission-driven, focused on transforming business processes through technology.
This role is part of an in-house Statement of Work (SOW) engagement. The client, end customer, and employer for this position is Bright Vision Technologies directly; there is no third-party client, vendor, or implementation partner involved. We do not engage in C2C, 1099, or third-party arrangements for this role.
About the Role
We are seeking an AI Performance Optimization Engineer to join our established and well-respected organization. This position offers tremendous career growth potential within a long-term, multi-year engagement aligned to our delivery roadmap.
In this role, you will focus on extracting maximum throughput, minimizing latency, and reducing cost across training and inference workloads for large neural network systems. The work spans the full stack, from low-level kernel optimization to distributed system tuning. You will require a deep understanding of GPU architecture, model parallelism, memory management, and compiler-level optimization.
The ideal candidate has demonstrated impact on production AI workloads, utilizing strong instrumentation and measurement discipline to enable rigorous, data-driven optimization decisions. You will work closely with cross-functional partners—including product, design, engineering, operations, and business stakeholders—to translate ambiguous requirements into well-engineered solutions. You will be expected to raise the bar through code review, design review, and mentorship of more junior engineers.
Hiring Process
For every role, a technical coding assessment is mandatory. Please apply only if you are confident in your technical abilities and hands-on experience.
To apply for immediate consideration, please send your resume to hilda@bvteck.com or contact us at (908) 698-4899. Learn more about Bright Vision Technologies at www.bvteck.com.
Equal Opportunity
Bright Vision Technologies is an Equal Opportunity Employer, including Disability/Veterans. We recognize that our people are our strength, and the diverse talents they bring to our global workforce are directly linked to our success.
We are committed to equal employment opportunity for all employees and applicants without regard to race, color, religion, sex, sexual orientation, gender identity or expression, national origin, age, genetic information, disability, veteran status, or any other protected status as defined by applicable federal, state, or local laws. This commitment extends to all aspects of employment, including recruitment, hiring, training, compensation, promotion, transfer, leaves of absence, termination, layoffs, and recall.
We expressly prohibit any form of workplace harassment or discrimination. We also make reasonable accommodations for applicants' and employees' religious practices and beliefs, as well as mental health or physical disability needs.
Work location
Work model: Remote
3155;3158, Cardenas Terrace, Centerville District, Fremont, Alameda County, California, 94536, United States
Fremont, California
Key Responsibilities
- check_circleProfile and optimize end-to-end AI training and inference pipelines for throughput, latency, and cost
- check_circleIdentify and eliminate bottlenecks across data loading, model compute, communication, and memory
- check_circleImplement and tune quantization, sparsity, and pruning strategies to reduce model footprint and accelerate inference
- check_circleOptimize distributed training using tensor parallelism, pipeline parallelism, FSDP, and ZeRO-style sharding
- check_circleTune attention implementations using FlashAttention, paged attention, and related techniques
- check_circleImplement KV cache optimization, continuous batching, and speculative decoding for LLM serving
- check_circleDrive compiler-level optimizations using Triton, XLA, TorchInductor, or TVM for measurable performance gains
- check_circleOptimize data pipelines, sharding strategies, and storage access patterns for high-throughput training
Requirements
- verifiedBachelor's or Master's degree in Computer Science, Computer Engineering, or related field
- verified6+ years experience in performance engineering, ML systems, or HPC
- verifiedStrong proficiency in Python and C++
- verifiedHands-on experience optimizing deep learning workloads on modern GPUs
- verifiedDeep understanding of distributed training and inference techniques
- verifiedExperience with profiling tools across CPU, GPU, and distributed systems
- verifiedFamiliarity with model compression techniques
- verifiedStrong grasp of memory hierarchies, communication primitives, and parallelism strategies
- verifiedNo new H1B sponsorship available; H1B transfers welcomed
Nice to Have
Experience optimizing LLM inference at production scale. Contributions to vLLM, TensorRT-LLM, DeepSpeed, or similar projects. Familiarity with custom kernel authoring in Triton or CUTLASS. Experience with FinOps for AI workloads. Publications or talks on AI systems performance.
