Lead the effort to make Runpod the fastest and most cost-efficient place for LLM inference, owning performance end to end.
Profile and diagnose performance bottlenecks across the serving stack, from scheduling to kernels, and implement fixes.
Work closely with product and infrastructure teams to shape how inference is offered, turning improvements into production-ready defaults.
Runpod is the AI Developer Cloud, providing a platform for over one million developers to experiment, train, fine-tune, deploy, and scale AI. We're a small, remote-first team that takes ownership seriously, moves fast, and has processed more than 20 billion inference requests.
Own and improve the AI inference serving stack, including prefill/decode disaggregation, continuous batching, KV-cache management, and speculative decoding.
Drive GPU performance engineering by identifying and eliminating bottlenecks using CUDA, Triton, profiling, and kernel optimization.
Set the serving and measurement roadmap, establish engineering standards, and mentor engineers while remaining hands-on with implementation.
Aleroy builds a unified platform for high-performance AI inference, enabling companies to deploy and run AI models at scale through a single API. We work in small, collaborative teams focused on solving complex engineering challenges in scalable AI infrastructure.
Own optimization of the inference stack: throughput, latency, cold starts, and cost per token.
Tune SGLang/vLLM engines, cut TTFT, boost KV cache reuse, and kill cold starts on 1.5 TB checkpoints.
Set technical direction with data, benchmark every change, and scale inference across nodes.
Kimchi is an LLM inference platform serving open models like DeepSeek and GLM from its own GPU fleet via one API line. It's made by CAST AI, a globally distributed team across 34 countries that hit unicorn status in 2026.
Bring new open-source models into production with a focus on latency and throughput.
Optimize inference systems including batching, KV cache, speculative decoding, and quantization.
Design, deploy, and operate heterogeneous clusters across infrastructure vendors to improve reliability and cost efficiency.
They are an early-stage AI infrastructure startup making large language model inference faster and more cost-efficient. The small team focuses on low-level performance engineering and AI-assisted workflows, with a strong emphasis on technical fundamentals.
Owns the architectural direction for Code Metal's AI platform and the team building it.
Builds production-grade inference serving, model gateways, agent harnesses, and context-engineering services.
Instruments the stack end to end and drives experimentation and observability for reliable AI systems.
Code Metal provides automated software engineering using AI and formal verification to produce provably correct code. Founded in 2023 and backed by Accel, Salesforce Ventures, and others, it serves defense and enterprise customers from offices in Boston and San Francisco.
Own end-to-end performance optimization for production ML workloads, from bottleneck identification to rollout.
Diagnose ML performance bottlenecks across model execution, frameworks, runtimes, memory, and CPU/GPU utilization.
Design and implement improvements in inference execution, runtime behavior, batching, memory management, and precision.
Reddit is a community of communities, hosting open and authentic conversations across the internet. With over 100,000 active communities and approximately 130 million daily active users, it is one of the web's largest sources of information.
Build inference systems powering RL research, scaling toward 1–3T parameter models on NVL72 clusters.
Optimize rollout throughput, multi-node MoE inference, and weight-update overhead for long-horizon agents.
Collaborate with model researchers on architecture choices balancing model quality and execution cost.
Prime Intellect is building the open superintelligence stack, unifying compute, environments, and deployment for frontier-scale AI post-training. Backed by $150M in funding, the team is mission-driven and hard-working, focused on accelerating open AI research.
Build production runtime infrastructure in Rust and Go for Modal's container platform.
Optimize container startup, checkpoint/restore, and snapshot pipelines for multi-GPU inference.
Diagnose complex Linux, kernel, and performance issues across the stack.
Modal builds serverless cloud infrastructure for AI/ML inference and training workloads. The company is an engineering-driven team that values deep systems work and open-source contributions.
Design, build, and evolve systems for distributed model training and batch inference at scale.
Develop workflows for experiment tracking, model evaluation, and registry management.
Optimize training and inference performance across GPU utilization, data loading, and distributed communication.
NOETIK is a frontier therapeutics company using AI to understand human biology and improve oncology outcomes, building foundation models on proprietary multimodal data. It fosters a collaborative, interdisciplinary culture focused on solving hard problems together.
Lead end-to-end technical deployments for GPU neocloud and AI Factory customers.
Configure and troubleshoot bare metal GPU infrastructure including CNI, GPU Operator, and storage.
Build playbooks and transfer knowledge to customer teams for self-sufficiency.
vCluster Labs provides a platform for AI infrastructure, helping GPU cloud providers and enterprises build Kubernetes-based clusters. They are a venture-backed startup with 40+ engineers and a remote-first culture, powering over 100,000 GPUs.