About the Role
The Turbo team operates at the intersection of efficient inference (algorithms, architectures, engines) and post‑training / RL systems. We develop and manage the systems powering Together’s API, including high‑performance inference and RL/post‑training engines capable of operating at production scale.
Our objective is to advance the frontier of efficient inference and RL‑driven training: making models significantly faster and more cost-effective to run, while simultaneously enhancing their capabilities through RL-based post-training (e.g., GRPO‑style objectives). This work involves both algorithms and systems: asynchronous RL, rollout collection, scheduling, and batching all interact with engine design, providing numerous parameters to adjust across the RL algorithm, training loop, and inference stack. A significant portion of the role involves modifying production inference systems—such as SGLang‑ or vLLM‑style serving stacks and speculative decoding systems like ATLAS—rooted in a strong understanding of post‑training and inference theory, rather than solely theoretical algorithm design.
You will work across the entire stack—from RL algorithms and training engines to kernels and serving systems—to develop and enhance frontier models using RL pipelines. Team members often exhibit specialized strengths: some are more proficient in RL, while others excel in systems. Depth in one of these areas, coupled with an eagerness to collaborate across disciplines (and grow towards more full‑stack ownership over time), is highly desirable.
Requirements
We do not expect every candidate to meet every single requirement. Team members typically possess deep expertise in one or more areas and sufficient breadth (or interest) to effectively work across the stack. The closer you are to a full-stack profile (inference + post-training/RL + systems), the stronger the fit—however, having deep expertise in one area and a strong desire to grow is perfectly acceptable.
You might be a good fit if you:
- Have strong expertise in at least one of the following, and are excited to collaborate across (and grow into) the others:
- Systems-first profile: Large-scale inference systems (e.g., SGLang, vLLM, FasterTransformer, TensorRT, custom engines, or similar), GPU performance, distributed serving.
- RL-first profile: RL / post-training for LLMs or large models (e.g., GRPO, RLHF/RLAIF, DPO-like methods, reward modeling), and using these to train or fine-tune real models.
- Model architecture design for Transformers or other large neural networks.
- Distributed systems / high-performance computing for ML.
- Are comfortable working from algorithms to engines:
- Strong coding ability in Python.
- Experience profiling and optimizing performance across GPU, networking, and memory layers.
- Able to take a new sampling method, scheduler, or RL update and transform it into a production-grade implementation within the engine and/or training stack.
- Have a solid research foundation in your area(s) of depth:
- Track record of impactful work in ML systems, RL, or large-scale model training (papers, open-source projects, or production systems).
- Can comprehend new RL / post-training papers, understand their implications on the stack, and design minimal, correct changes in the appropriate layer (training engine vs. inference engine vs. data / API).
- Operate well as a full-stack problem solver:
- You naturally ask: “Where in the stack is this really bottlenecked?”
- You enjoy collaborating with infra, research, and product teams, and you value both scientific quality and user-visible achievements.
Minimum qualifications
- 3+ years of experience working on ML systems, large-scale model training, inference, or adjacent areas (or equivalent experience via research / open source).
- Advanced degree in Computer Science, EE, or a related field, or equivalent practical experience.
- Demonstrated experience owning complex technical projects end-to-end.
If you’re excited about the role and strong in some of these areas, we encourage you to apply even if you don’t meet every single requirement.
Responsibilities
- Advance inference efficiency end-to-end
- Design and prototype algorithms, architectures, and scheduling strategies for low-latency, high-throughput inference.
- Implement and maintain changes in high-performance inference engines (e.g., SGLang- or vLLM-style systems and Together’s inference stack), including kernel backends, speculative decoding (e.g., ATLAS), quantization, etc.
- Profile and optimize performance across GPU, networking, and memory layers to improve latency, throughput, and cost.
- Unify inference with RL / post-training
- Design and operate RL and post-training pipelines (e.g., RLHF, RLAIF, GRPO, DPO-style methods, reward modeling) where 90+% of the cost is inference, jointly optimizing algorithms and systems.
- Make RL and post-training workloads more efficient with inference-aware training loops—for example, async RL rollouts, speculative decoding, and other techniques that make large-scale rollout collection and evaluation cheaper.
- Use these pipelines to train, evaluate, and iterate on frontier models on top of our inference stack.
- Co-design algorithms and infrastructure so that objectives, rollout collection, and evaluation are tightly coupled to efficient inference, and quickly identify bottlenecks across the training engine, inference engine, data pipeline, and user-facing layers.
- Run ablations and scale-up experiments to understand trade-offs between model quality, latency, throughput, and cost, and feed these insights back into model, RL, and system design.
- Own critical systems at production scale
- Profile, debug, and optimize inference and post-training services under real production workloads.
- Drive roadmap items that require real engine modification—changing kernels, memory layouts, scheduling logic, and APIs as needed.
- Establish metrics, benchmarks, and experimentation frameworks to validate improvements rigorously.
- Provide technical leadership (Staff level)
- Set technical direction for cross-team efforts at the intersection of inference, RL, and post-training.
- Mentor other engineers and researchers on full-stack ML systems work and performance engineering.