Forward Deployed Engineer, Post-Training

Forward Deployed Engineer, Post-Training

Together AI is building the AI Native Cloud, a platform purpose-built for AI engineers who need high-performance inference, fine-tuning, reinforcement learning capabilities, and pre-training at scale. The company provides access to a marketplace of leading open-source models and serves over 400 trillion tokens monthly to customers including Cursor, Decagon, ElevenLabs, Salesforce, and Zoom.

Together AI's platform enables teams to run, adapt, and own their own models. As the infrastructure layer for AI development, the company addresses a critical gap in the market by offering the tools and compute resources that developers need to build specialized, frontier-level models. This is especially relevant as AI teams increasingly recognize that fine-tuning and reinforcement learning on open-source foundations can deliver significant competitive advantages over relying on closed APIs alone.

In this Forward Deployed Engineer role on the Customer Success team, you will be the deep technical expert on post-training for our most strategic customers. You will work alongside Solutions Architects as a hands-on partner, helping AI teams navigate data strategy, algorithm selection, experiment design, and compute sizing as they build differentiated models through fine-tuning and reinforcement learning. You will own the technical execution of post-training proofs of concept, drive platform adoption, and ensure customer goals translate into measurable improvements in model quality.

What you'll do

• Partner with customers to define data strategies for model post-training, identify data quality bottlenecks, and recommend approaches for generating training data at scale

• Evaluate different post-training methodologies and recommend the optimal approach for each customer's unique use case and objectives

• Design controlled experiments with proper tracking and evaluation frameworks, diagnose training issues like reward hacking and performance regressions, and run rigorous ablations to separate real gains from noise

• Build evaluation systems alongside customers, including task-specific benchmarks, LLM-as-judge scoring pipelines, and held-out test sets that validate improvements against production metrics

• Assess open-source base models, guide model selection methodology, and establish baseline quality evaluations to inform post-training objectives

• Size reinforcement learning and fine-tuning workloads, partner with sales and capacity teams to shape customer deals and forecast infrastructure needs

• Serve as the primary technical contact for strategic accounts, leading execution from initial scoping through production deployment and milestone achievement

• Feed insights from customer work back into the product roadmap, contribute code and reference implementations, and drive early adoption of new features

What you'll bring

• Five or more years of hands-on technical experience in machine learning engineering, applied AI, forward deployment, or solutions architecture, with recent direct experience post-training large language models

• Expert-level, hands-on proficiency with fine-tuning and reinforcement learning for LLMs, including techniques such as SFT, LoRA, DPO, GRPO, and PPO

• Proven ability to design rigorous experiments, run meaningful ablations, build robust evaluations, and distinguish genuine improvements from experimental noise

• Demonstrated success building, iterating, and refining training datasets to achieve model performance goals

• Comprehensive knowledge of current open-source model landscape with strong judgment about which base models suit specific customer requirements

• Strong Python skills and comfort working within production environments

• Track record leading complex, cross-functional technical projects with exceptional communication and project management abilities

What they offer

• Base salary range of 270,000 to 300,000 USD plus equity and benefits

• Health insurance coverage

• Flexibility around remote work arrangements

• Startup equity participation