AI Research Engineer (Kernel & Inference Optimization) - 100% Remote Worldwide
Tether Operations Limited1 month ago
Remote, WorldwideSenior
Responsibilities
- Design and deploy model-serving architectures optimized for throughput, latency, memory usage, mobile devices, and edge platforms.
- Build, run, and monitor inference tests in simulated and production environments using metrics such as latency, throughput, memory consumption, and error rates.
- Create test datasets, simulation scenarios, benchmarks, and measurable evaluation criteria for resource-constrained deployments.
- Analyze processing and memory metrics to identify bottlenecks involving batching, network delays, and high memory usage.
- Develop and integrate optimized inference frameworks and pipelines for edge and on-device applications.
- Optimize distributed inference engines using tensor parallelism, pipeline parallelism, and expert parallelism for GPU clusters.
- Collaborate with cross-functional teams and continuously refine systems based on production performance data.
Requirements
- Bachelor’s degree in Computer Science or a related field is required; a PhD in NLP, Machine Learning, or a related field is preferred.
- A strong track record in AI research and development, ideally including publications at A* conferences, is preferred.
- Experience with Metal Shading Language and writing custom compute shaders from scratch is required.
- Proven experience with low-level kernel optimization and inference optimization on mobile devices is required.
- Demonstrated improvements in inference latency, throughput, and memory footprint for domain-specific applications are required.
- Deep knowledge of model-serving architectures, inference frameworks, inference engines, and efficient memory-management techniques is required.
- Strong expertise writing GPU kernels for smartphones and other mobile devices is required.
- Practical experience developing and deploying end-to-end inference pipelines on resource-constrained devices and edge platforms is required.
- Experience applying empirical research to latency optimization, computational bottlenecks, and memory constraints is required.
- Proficiency designing evaluation frameworks and iterating on inference optimization strategies is required.
- Understanding of Diffusion Models, Vision Transformers, pruning, quantization, FlashAttention, KV cache, and speculative decoding such as Eagle is required.
- Strong English communication skills are expected.
Benefits
- 100% remote work worldwide.
- Opportunity to work with a global, distributed team in digital finance, AI, and peer-to-peer technology.
Categories
AI ResearchML Engineering
About Tether Operations Limited
Tether Operations Limited develops and issues reserve-backed digital tokens used by exchanges, wallets, payment processors, and merchants, most notably the USDT stablecoin available across multiple blockchains. Founded in 2014 and privately held, it earns revenue from token issuance, reserve management, and related services, and has expanded into tokenization, peer-to-peer communications (Keet), and Bitcoin mining. USDT is widely used in crypto trading and cross-border transfers, including in emerging markets.