Applied AI/ML Engineer - End-to-End GPU Inference & RL

Carbon Data Solutions; United States; Remote; USD 175,000 - 250,000; Full time

Benefits offered by this job

Equity allocation

Health/dental/vision (US)

Flexible PTO

Conference travel budget

Remote-first with off-sites

Job summary

Boundless is seeking an Applied AI/ML Engineer to ship AI-powered products end-to-end on our growing GPU inference fleet. You will own features from prototype to production, optimize throughput and cost on real GPUs, and iterate rapidly based on feedback.

You will flex across the stack, teaming with Infra on GPU scheduling and with Product on what to build next and why. A strong bias for action and comfort with ambiguity are essential.

Qualifications

• 3+ years shipping ML/AI systems to production.
• Hands-on experience serving LLM inference with vLLM, SGLang, or TensorRT-LLM.
• Experience with RL / post-training methods (GRPO, PPO, DPO, or SFT), or strong adjacent experience and a clear desire to go deep here.
• Strong Python and PyTorch.
• Working understanding of GPU execution: batching, memory, and basic CUDA concepts.
• Comfort operating in ambiguity with a strong bias for action.

Responsibilities

• End-to-End AI Product Delivery: Own AI features from prototype through production, including model selection, serving, evaluation, and iteration.
• Inference Serving: Deploy and optimize LLM inference across the fleet using vLLM and SGLang; tune batching, KV-cache, quantization, decoding, and multi-model routing.
• RL & Post-Training Harnesses: Build and operate reinforcement-learning and post-training pipelines with Megatron-LM + SGLang and Prime Intellect, including reward design and stability.
• Evaluation & Iteration: Build eval harnesses and benchmarks to measure quality, throughput, and cost; drive data-informed iteration.
• Work Across the Stack: Partner with Infrastructure on GPU scheduling and with Product on roadmap decisions.

Skills

ML in prod

Python

PyTorch

LLM inference

vLLM

SGLang

TensorRT-LLM

RL methods

CUDA basics

Bias for action

Tools

Megatron-LM

verifiers library

slime

prime-rl

Ray

SkyPilot

Slurm

Job description

Boundless is seeking an Applied AI/ML Engineer to ship AI-powered products end-to-end on our growing GPU inference fleet. You will own features from prototype to production, optimize throughput and cost on real GPUs, and iterate rapidly based on feedback.

You will flex across the stack, teaming with Infra on GPU scheduling and with Product on what to build next and why. A strong bias for action and comfort with ambiguity are essential.