Research Engineer, Training & Inference

Harmonic Security; Palo Alto (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Unlimited PTO

401(k) matching

100% employer-paid health, vision, and

dental benefits for employees and 50%

Job summary

Harmonic Security is transforming AI with a math-first reinforcement learning stack. We seek engineers to own and optimize the full RL training and serving pipeline, from Python APIs to CUDA kernels, to reach peak throughput on foundation models.

You will drive distributed training across multi-node GPU clusters, tailor inference for high-throughput RL and LLM workloads, and push memory efficiency while balancing compute-heavy training cycles.

Qualifications

• BS in Computer Science or a related technical field, or equivalent industry experience.
• 2+ years of relevant, hands-on industry experience.
• Proficiency in Python.
• Experience building or maintaining components within ML frameworks (e.g., PyTorch, JAX, or TensorFlow).
• Proficiency in either: Understanding of distributed training concepts and collective communication primitives (e.g., NCCL). Practical experience deploying and profiling models on GPU-accelerated cloud infrastructure.

Responsibilities

• Total Stack Ownership: Maintain and optimize our proprietary RL training and serving infrastructure; refactor layers from Python API to CUDA kernels for peak performance.
• Optimized Training: maximize throughput of RL data generation to model training with sharded multi-node training and inference.
• High-Performance Serving: optimize inference stack for high-throughput RL and low-latency LLM production traffic; tune engine, router, and scheduler; implement kernels if needed.
• Compute Optimization: identify bottlenecks in distributed clusters to maximize throughput and memory efficiency for multi-billion parameter models.

Skills

Python

ML frameworks (PyTorch/JAX/TensorFlow)

Distributed training concepts

GPU cloud infra

Education

BS in Computer Science or related

MS/PhD in CS/Math or related

Tools

CUDA

C++

Triton

CuTeDSL

TileLang

CUTLASS

ThunderKittens

Job description

About Harmonic

At Harmonic, we are building a mathematical reasoning engine that operates with absolute precision. While most AI makes maximum-likelihood guesses, Harmonic's Aristotle uses Lean 4 and reinforcement learning to verify its reasoning and results.
Following our Gold Medal-level performance on the 2025 International Math Olympiad (IMO) and the successful resolution of long-standing open problems, we are proving that AI can master the most rigorous domains of human thought. Backed by some of the world’s most prominent investors, we are intentionally scaling an elite technical team.
Visit our company blog to learn more about what we are working on!

About the Role

We are developing reinforcement learning systems at a scale where standard abstractions frequently fail. Unlike labs that operate primarily through high-level wrappers, we own the entirety of our RL stack. This ownership spans from low-level environment simulators and custom communication primitives to our distributed training loops and inference engines.

We are seeking engineers who view existing libraries as a baseline and the hardware speed itself as the true target. You will be responsible for the architecture powering our agents, with a relentless focus on maximizing the throughput of our reinforcement learning and production workflows.

Key Responsibilities

• Total Stack Ownership: Maintain and optimize our proprietary RL training and serving infrastructure. You have the authority to refactor any layer—from the Python API down to the CUDA kernels—to achieve peak performance for foundation model workloads.

• Optimized Training: maximize the throughput of our reinforcement learning system from data generation to model training with sharded multi-node training and inference algorithms.

• High-Performance Serving: optimize our inference stack for high-throughput reinforcement learning and low-latency LLM production traffic. Tune the inference engine, router, and scheduler, down to custom kernels if need be.

• Compute Optimization: Identify and resolve performance bottlenecks within our distributed clusters, ensuring optimal throughput and memory efficiency for multi-billion parameter models, balancing memory constraints with compute-heavy training cycles.

Minimum Qualifications

• BS in Computer Science or a related technical field, or equivalent industry experience

• 2+ years of relevant, hands‑on industry experience

• Proficiency in Python

• Experience building or maintaining components within ML frameworks (e.g., PyTorch, JAX, or TensorFlow).

• Proficiency in either:

  • Understanding of distributed training concepts and collective communication primitives (e.g., NCCL).

  • Practical experience deploying and profiling models on GPU-accelerated cloud infrastructure.

Preferred Qualifications

• MS or PhD in Computer Science, Mathematics, or a related field.

• 5+ years of relevant, hands‑on industry experience

• Proficiency in C++

• Experience writing or improving kernels (Triton, CuTeDSL, TileLang, CUDA, CUTLASS, ThunderKittens) to resolve low-level bottlenecks.

• Proven success deploying performant inference at scale using open-source or custom inference engines, routers, etc.

• Direct experience scaling models via FSDP, Tensor Parallelism, or related sharding techniques on multi-node GPU clusters.

• Experience designing reinforcement learning systems for high-throughput training and asynchronous data sampling.

What We Offer

• Unlimited PTO

• 401(k) matching

• 100% employer-paid health, vision, and dental benefits for employees and 50% coverage for dependents. Harmonic offers varied health coverage options to select what is best for you and your family.

• Health Savings Account (HSA) available for qualifying health plans

Equal Opportunity Statement

Harmonic is committed to diversity and inclusivity in the workplace. We are an equal opportunity employer and do not discriminate on the basis of race, religion, national origin, gender, sexual orientation, age, veteran status, disability or any other legally protected status.