Remote AI Inference & Post-Training Specialist

Togetherai; San Francisco (CA); On-site; USD 270,000 - 300,000; Full time

Benefits offered by this job

Startup equity

Health insurance

Remote work flexibility

Job summary

Together AI, the AI Native Cloud, seeks a Forward Deployed Engineer to partner with strategic customers on inference optimization, post-training pipelines, and production deployment. You will drive hands-on tuning, guide LoRA/SFT/DPO/RLHF/GRPO pipelines, and align configurations with milestones, delivering high-impact results.

Join a team that values field insights, platform adoption, and collaborative product feedback to harden the core AI platform while enabling scalable, high-performance

Qualifications

• 5+ years in a technical role with focus on inference systems or post-training workflows.
• Hands-on experience with inference engines (vLLM, TensorRT-LLM, SGLang).
• Strong Python skills and production deployment experience.
• Experience with fine-tuning, LoRA, SFT, DPO, RLHF, and GRPO workflows.

Responsibilities

• Optimize inference engines for hardware, model architecture, and workload profiles.
• Develop configuration updates to win POCs, benchmarks, and deployments; tune caches and parallelism.
• Lead post-training and fine-tuning, guiding LoRA, SFT, DPO, RLHF, and GRPO pipelines.
• Act as a technical contact for strategic accounts, aligning configurations with milestones.
• Onboard customers with correct inference and post-training setups from day one.
• Provide feedback to product roadmap based on field insights and adopter patterns.

Skills

Inference optimization

Python

Post-training pipelines

Open-source deployment

Tools

vLLM

TensorRT-LLM

SGLang

Job description

Together AI, the AI Native Cloud, seeks a Forward Deployed Engineer to partner with strategic customers on inference optimization, post-training pipelines, and production deployment. You will drive hands-on tuning, guide LoRA/SFT/DPO/RLHF/GRPO pipelines, and align configurations with milestones, delivering high-impact results.

Join a team that values field insights, platform adoption, and collaborative product feedback to harden the core AI platform while enabling scalable, high-performance