Pika
ML Engineer, Inference & Optimization
US remote. Posted 4 months ago
The role
You will design and optimize inference pipelines and implement acceleration techniques. You will work on GPU parallelism and high-performance computing kernels to power video and language models.
What you will do
- Lead and implement advanced inference acceleration techniques including attention optimization and quantization.
- Engineer and optimize GPU strategies across tensor, sequence, and pipeline parallelism.
- Develop and optimize high-performance computing kernels and distributed workloads using CUDA and NCCL.
- Collaborate with research and engineering teams to bring videogen and large language models into production.
- Drive rigorous code reviews, participate in technical discussions, and mentor fellow engineers.
Required qualifications
- 5+ years engineering experience with a track record in inference acceleration and model deployment at scale.
- Proven expertise in inference optimization including quantization, attention acceleration, and deep learning compiler stacks.
- Deep knowledge of GPU programming (CUDA, NCCL).
- Experience with SP, TP, PP, and other forms of parallelism for distributed inference.
- Familiarity with video generation (videogen) models and large language models (LLMs).
- Strong cross-discipline communication skills.
- Self-driven and capable of managing ambiguity in a fast-paced startup environment.
Nice to have
- Experience in enhancing training efficiency, stability, or resource optimization for large models.
- Experience with high-throughput video or real-time streaming model deployment.
- Familiarity with distributed training and optimization toolkits.
- Contributions to open source projects in AI infrastructure or deep learning compilers.
- Startup or rapid prototyping experience.
Benefits
- Competitive salary in the AI industry
- Equity in a fast-growing startup
- Comprehensive health benefits
- Monthly stipends
- Company retreats
About the company
Pika is a tight-knit, energetic team based in Palo Alto, CA, working to make video creation seamless and intuitive using AI.