---
title: 'Research Engineer - AI Performance & Kernel Optimization at Zyphra'
canonical: 'https://feeny.ai/job/research-engineer-ai-performance-kernel-optimization-zyphra-san-francisco-tgaps2t9x56z'
type: 'job'
last_seen: '2026-09-06'
---

# Research Engineer - AI Performance & Kernel Optimization at Zyphra

- **Company:** Zyphra
- **Location:** San Francisco, CA
- **Employment:** full-time
- **Work type:** onsite
- **Posted:** 2026-03-16
- **Last confirmed live:** 2026-09-06
- **Apply:** https://jobs.ashbyhq.com/zyphra/c94162cf-d121-43cc-a17f-afb54ae000e8

## Job description

ZYPHRA IS AN ARTIFICIAL INTELLIGENCE COMPANY BASED IN SAN FRANCISCO, CALIFORNIA.

## THE ROLE:

As a Research Engineer - AI Performance & Kernel Optimization, you will improve and optimize the performance of our large-scale language model training and inference stacks. You will work closely with our pretraining and inference teams to identify bottlenecks, design and implement highly optimized kernels, and push the limits of throughput, latency, and hardware utilization across a range of accelerator platforms. This role is suited for someone who enjoys deep systems work, cares about performance at every level of the stack, and is excited to translate low-level optimizations into meaningful gains for frontier-scale AI systems.

## YOU’LL WORK ACROSS:

- Kernel development and optimization for large-scale ML workloads, using any level of the stack from PTX/assembly to CUDA, HIP, Triton, or other GPU DSLs
- Performance tuning for training and inference stacks across GPUs and other accelerators
- Profiling and eliminating bottlenecks in memory movement, communication, scheduling, and compute utilization
- Optimizing distributed training and inference systems for large MoE models, including large-scale model parallelism
- Portability and optimization across non-NVIDIA hardware, with special interest in AMD hardware such as the MI300x and MI355x
- Collaboration with research and infrastructure teams to turn systems improvements into real-world model training and inference gains

WHAT WE'RE LOOKING FOR / REQUIREMENTS:

- Strong engineering aptitude for building reliable, high-performance systems
- Excellent low-level performance intuition and the ability to reason about hardware-software interactions
- Are excited to rapidly learn new systems, tools, and hardware environments
- Excellent communication and collaboration skills, with the ability to work effectively across research and engineering teams
- Enjoy diving deep into the weeds and hunting down the last 10–20% of performance

## QUALIFICATIONS / ADDITIONAL SKILLS:

- Experience writing highly performant GPU kernels at any level of abstraction–PTX, CUDA, HIP, Triton, or other kernel DSLs
- Experience optimizing ML workloads for large-scale training, ideally in language model pretraining or inference environments
- Experience with non-NVIDIA accelerator hardware, such as AMD, AWS Trainium, Google TPU, Qualcomm, ARM, Intel, and custom ASICs
- Strong understanding of distributed training systems and parallelism schemes, including data parallelism, tensor/model parallelism, pipeline parallelism, sharding, and communication/computation overlap
- Experience with performance engineering in other demanding parallel computing environments such as HPC, quantitative finance, scientific computing, graphics, compilers, or numerical simulation
- Strong systems intuition around memory hierarchy, bandwidth constraints, kernel fusion, launch overhead, communication overhead, and hardware utilization
- Experience using profiling and debugging tools to drive performance improvements
- Familiarity with infrastructure underlying large-scale training and inference, including collective communication libraries, and runtime performance analysis
- Background in a highly technical field such as physics, mathematics, theoretical computer science, computer science, or electrical engineering
- Any HPC experience is a strong plus

## WHY WORK AT ZYPHRA:

- Our research methodology is grounded in methodical, step-by-step approaches to ambitious goals. Both deep research and engineering excellence are equally valued
- We strongly value new and crazy ideas and are very willing to bet big on new ideas
- We move as quickly as we can; we aim to minimize the bar to impact as low as possible
- We all enjoy what we do and love discussing AI

## BENEFITS AND PERKS:

- Comprehensive medical, dental, vision, and FSA plans
- Competitive compensation and 401(k) plan
- Relocation and immigration support on a case-by-case basis
- In-office snacks and meals provided
- Unlimited PTO and company holidays
- In-person team in San Francisco with a collaborative, high-energy environment

## About Zyphra

## Company Overview
- **One-liner**: Zyphra is a full-stack AI company building open foundation models and a sovereign cloud platform to deliver advanced AI to developers, enterprises, and hyperscalers.
- **Entity Type**: Private, Series A
- **Headquarters**: San Francisco, California, United States
- **Founded**: 2020 (some sources cite 2021)
- **Founders**: Krithik Puthalath (CEO), Beren Millidge (Chief Scientist), Danny Martinelli (Head of Product), Tomás Figliolia (Chief Architect)

## Core Business
- **Primary industry**: Artificial Intelligence / Foundation Models / Cloud AI Infrastructure
- **Target customers**: Developers, enterprises, and "frontier AI hyperscalers" (B2B)
- **Mission or purpose**: "Open Superintelligence" – to enable sovereign, transparent, and customizable AI that runs on any hardware without vendor lock-in.

## Products & Services
- **[Zyphra Research](https://www.zyphra.com/)**: Open-source foundation models trained on heterogeneous compute, focusing on novel architectures for long-term memory, continual learning, and silicon performance. Released as research prototypes.
- **[Zyphra Cloud](https://www.zyphra.com/)**: Full-stack AI platform built on AMD, designed for long-horizon agents. Provides managed infrastructure for deploying, customizing, and running Zyphra's or third-party models with full data sovereignty.

## Market Standing
- **Valuation/Market Cap**: Not disclosed
- **Key Metric**: Total funding of **$11M** (Series A)
- **Notable Investors/Partners**: AMD, Intel Capital, Future Ventures, Bison Ventures, Transpose Platform, and others
- **Growth Signals**:
  - Headcount grew **151.9% YoY** to 53 employees
  - Operates in 5 countries (US, UK, Canada, India, Saudi Arabia) with offices in San Francisco, Palo Alto, and London
  - Active job postings: 14 (up 16.7% quarterly)
  - LinkedIn followers grew 236.3% year-over-year

## Competitive Advantages
- **Full-stack open-source approach**: Combines novel model architectures (probabilistic computing, heterogeneous compute) with a cloud platform optimized for long-horizon AI agents.
- **Hardware agnosticism**: Promises "run intelligence across any hardware and any provider, without lock-in," with early focus on AMD silicon.
- **Commitment to openness**: Models are transparent, allowing customers to audit reasoning and adapt to specific domains.
- **Sovereignty focus**: Enterprises own their models, data, and deployments – appealing to regulated industries.

## Strategic Focus
- **Priority on architecture innovation** (continual learning, long-term memory, silicon performance) over scaling existing methods.
- **Building a full-stack platform** that tightly couples research breakthroughs with a production-grade cloud service.
- **Scaling the team** aggressively – hiring across research engineering, platform, developer relations, and AI engineering.

## Why Work Here
- **Culture of open research**: Employees contribute to novel model architectures that are publicly released, offering a rare blend of academic freedom and product impact.
- **Heavy engineering focus**: 57% of the team is in technical roles, with a flat structure (62% specialist-level roles).
- **Locations**: San Francisco (HQ) and London, with potential for remote-friendly roles (not explicitly stated).
- **Notable perks**: Not detailed publicly, but typical for a well-funded Series A startup (equity, benefits, etc.).
- **Growth trajectory**: 150%+ headcount growth signals rapid scaling and career advancement opportunities.

## Sources
1. [zyphra.com](https://www.zyphra.com/)
2. [zyphra.com about](https://www.zyphra.com/about)
3. [jobs.ashbyhq.com/zyphra](https://jobs.ashbyhq.com/zyphra)
4. [linkedin.com/company/zyphra](https://www.linkedin.com/company/zyphra)
5. [cbinsights.com/company/zyphra](https://www.cbinsights.com/company/zyphra)

## Other roles at Zyphra

- [GTM Engineer](https://feeny.ai/job/gtm-engineer-zyphra-san-francisco-036cvpkp3amk) — San Francisco, CA
- [Principal Solutions Architect](https://feeny.ai/job/principal-solutions-architect-zyphra-san-francisco-tvmr7yr66sye) — San Francisco, CA
- [Developer Relations Lead](https://feeny.ai/job/developer-relations-lead-zyphra-san-francisco-8aak96cd9gcr) — San Francisco, CA
- [Research Engineer - Model Architectures](https://feeny.ai/job/research-engineer-model-architectures-zyphra-san-francisco-q2xft42gbdne) — San Francisco, CA
- [Research Engineer - Agency and Reasoning](https://feeny.ai/job/research-engineer-agency-and-reasoning-zyphra-san-francisco-n78mrpw38xb2) — San Francisco, CA
- [Research Engineer - Language Model Pre-Training](https://feeny.ai/job/research-engineer-language-model-pre-training-zyphra-san-francisco-1e2n0tewvz0d) — San Francisco, CA
- [Research Engineer - Audio & Speech Models](https://feeny.ai/job/research-engineer-audio-speech-models-zyphra-san-francisco-vpqsx2qaz8pm) — San Francisco, CA
- [Data Engineer - Multimodal Systems](https://feeny.ai/job/data-engineer-multimodal-systems-zyphra-san-francisco-kd4wkewajppq) — San Francisco, CA
- [Platform Engineer](https://feeny.ai/job/platform-engineer-zyphra-san-francisco-7nkavn90wn8m) — San Francisco, CA
- [Lead Frontend Engineer](https://feeny.ai/job/lead-frontend-engineer-zyphra-san-francisco-d8cyqx60q9fe) — San Francisco, CA