---
title: 'Distributed Training and Inference Engineer at Sciforium'
canonical: 'https://feeny.ai/job/distributed-training-and-inference-engineer-sciforium-san-francisco-13qdzh34vrwv'
type: 'job'
last_seen: '2026-09-13'
---

# Distributed Training and Inference Engineer at Sciforium

- **Company:** Sciforium
- **Location:** San Francisco, CA
- **Employment:** full-time
- **Work type:** onsite
- **Posted:** 2026-08-24
- **Last confirmed live:** 2026-09-13
- **Apply:** https://jobs.ashbyhq.com/sciforium/1471adc1-cf58-4174-8dc2-e8c0e29cfb17

## Job description

Sciforium is an AI infrastructure company developing next-generation multimodal AI models and a proprietary, high-efficiency serving platform. Backed by multi-million-dollar funding and direct sponsorship from AMD with hands-on support from AMD engineers the team is scaling rapidly to build the full stack powering frontier AI models and real-time applications.

## About the role

Sciforium is seeking a highly skilled Distributed Training and Inference Engineer to build, optimize, and maintain the critical software stack that powers our large-scale AI training and serving workloads. In this role, you will work across the entire machine learning infrastructure from low-level CUDA/ROCm runtimes to high-level frameworks like JAX and PyTorch to ensure our distributed training systems are fast, scalable, stable, and efficient. This position is ideal for someone who loves deep systems engineering, debugging complex hardware–software interactions, and optimizing performance at every layer of the ML stack. You will play a pivotal role in enabling the training and deployment of next-generation LLMs and generative AI models.

## What you'll do

- Software Stack Maintenance: Maintain, update, and optimize critical ML libraries and frameworks including JAX, PyTorch, CUDA, and ROCm across multiple environments and hardware configurations.
- End-to-End Stack Ownership: Build, maintain, and continuously improve the entire ML software stack from ROCm/CUDA drivers to high-level JAX/PyTorch tooling.
- Distributed System Optimization: Ensure all model implementations are efficiently sharded, partitioned, and configured for large-scale distributed training and serving.
- System Integration: Continuously integrate and validate modules for runtime correctness, memory efficiency, and scalability across multi-node GPU/accelerator clusters.
- Profiling & Performance Analysis: Conduct detailed profiling of compilation graphs, training workloads, and runtime execution to optimize performance and eliminate bottlenecks.
- Debugging & Reliability: Troubleshoot complex hardware–software interaction issues, including vLLM compilation failures on ROCm, CUDA memory leaks, distributed runtime failures, and kernel-level inconsistencies.
- Collaborate with research, infrastructure, and kernel engineering teams to improve system throughput, stability, and developer experience.

Ideal candidate profile

- 5+ years of industry experience in ML systems, distributed training, or related fields.
- Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or related technical fields.
- Strong programming experience in Python, C++, and familiarity with ML tooling and distributed systems.
- Deep understanding of profiling tools (e.g., Nsight, ROCm Profiler, XLA profiler, TPU tools).
- Deep expertise with partitioning configuration on the modern ML frameworks such as PyTorch and JAX.
- Experience with multi-node distributed training systems and orchestration frameworks (DTensor, GSPMD, etc.).
- Hands-on experience maintaining or building ML training stacks involving CUDA, ROCm, NCCL, XLA, or similar technologies.

Nice-to-have

- Extensive experience with the XLA/JAX stack, including compilation internals and custom lowering paths.
- Familiarity with distributed serving or large-scale inference frameworks (e.g., vLLM, TensorRT, FasterTransformer).
- Background in GPU kernel optimization or accelerator-aware model partitioning.
- Strong understanding of low-level C++ building blocks used in ML frameworks (e.g., XLA, CUDA kernels, custom ops).

## Benefits include

- Medical, dental, and vision insurance
- 401k plan
- Daily lunch, snacks, and beverages
- Flexible time off
- Competitive salary and equity

## Equal opportunity

Sciforium is an equal opportunity employer. All applicants will be considered for employment without attention to race, color, religion, sex, sexual orientation, gender identity, national origin, veteran or disability status.

## About Sciforium

## Company Overview
- **One-liner**: Sciforium is building a vertically integrated AI infrastructure platform that owns its hardware (AMD GPUs) to deliver cost-effective, high-performance inference and foundation model training across text, image, video, and audio modalities.
- **Entity Type**: Private (Seed Stage)
- **Headquarters**: San Francisco, California, United States
- **Founded**: 2024
- **Founders**: Hassan Akbari

## Core Business
- **Primary Industry**: AI Infrastructure / Generative AI
- **Target Customers**: B2B; AI teams and enterprises that need scalable, multimodal AI inference and model serving without managing their own infrastructure.
- **Mission**: To rebuild AI serving infrastructure from the ground up—owning the hardware and optimizing the entire pipeline end to end—so that any team, regardless of size or budget, can access the best AI capabilities across every modality without compromise.

## Products & Services
- **AI Inference API**: A drop-in replacement for the OpenAI API format, supporting streaming, tool use, structured outputs, and async generation. Runs on Sciforium’s own AMD hardware for lower cost and stronger privacy.
- **Evaluation Platform**: Built-in pipelines to monitor model performance in real time, catch regressions, and benchmark across models.
- **Native Agents Infrastructure**: Serverless platform for running AI agents at scale without managing servers.
- **Model Library**: Access to state-of-the-art open-source models across text, image, video, and audio (e.g., DeepSeek, Wan2, speech models).

## Market Standing
- **Valuation/Market Cap**: Not disclosed
- **Key Metric**: Total Funding of $3.9M (Seed round closed June 2024)
- **Notable Investors/Partners**: Backed by AMD and SignalFire
- **Growth Signals**: Headcount grew 160% YoY to 8 employees; 10 active job postings as of mid-2025; 99.98% uptime claimed; N+1 power and N+2 cooling redundancy with liquid-cooled infrastructure.

## Competitive Advantages
- **Vertical Integration**: Owns its own AMD hardware and runs its own data centers, removing intermediaries and markups—leading to lower costs and predictable performance.
- **Multimodal Native**: Built from the ground up to handle text, image, video, and audio in a single API, unlike many competitors that focus on text-only.
- **Privacy & Control**: By running on dedicated infrastructure (not shared servers), customers get stronger data privacy guarantees.
- **High Ambition Culture**: Team includes alumni from Google DeepMind, Microsoft, Amazon, Snowflake, Qualcomm, and Columbia University.

## Strategic Focus
- **Infrastructure Ownership**: Continuing to invest in its own AMD GPU clusters and data center operations to maintain cost and performance advantages.
- **Multimodal Expansion**: Scaling support for all data types (text, image, video, audio) with a single API.
- **Agent Readiness**: Building native support for AI agents at scale.
- **Open-Source Ecosystem**: Supporting the latest open-source models on day one.

## Why Work Here
- **Culture**: Highly independent, self-motivated, and creative environment. Small enough that your work is visible from day one. Principles include relentless quality, outcome ownership, and high ambition.
- **Work Policy**: Hybrid and in-office roles available. Offices in San Francisco (HQ) and Los Altos, California.
- **Team**: Small, high-caliber team with deep experience from Google DeepMind, Snowflake, Amazon, Qualcomm, and other top AI/infra companies.
- **Perks**: Work on hard infrastructure problems that matter, at a company backed by AMD and SignalFire. Opportunity to shape the foundation of AI infrastructure from an early stage.

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

## Other roles at Sciforium

- [Data Center Real Estate & Development Specialist](https://feeny.ai/job/data-center-real-estate-development-specialist-sciforium-san-francisco-1mzk59gk4ad1) — San Francisco, CA
- [GPU Kernel Engineer](https://feeny.ai/job/gpu-kernel-engineer-sciforium-san-francisco-95qxvmm7nxrh) — San Francisco, CA
- [GPU Kernel Engineer](https://feeny.ai/job/gpu-kernel-engineer-sciforium-san-francisco-xm87qw2qrrt1) — San Francisco, CA
- [Pre-training Research Engineer](https://feeny.ai/job/pre-training-research-engineer-sciforium-san-francisco-qqxzt72sswez) — San Francisco, CA
- [Research Engineer - Model Evaluation & MLOps](https://feeny.ai/job/research-engineer-model-evaluation-mlops-sciforium-san-francisco-j6bk6vmfyth9) — San Francisco, CA
- [Lead Software Engineer, Model Serving Platform](https://feeny.ai/job/lead-software-engineer-model-serving-platform-sciforium-san-francisco-sbdhk82sxhx3) — San Francisco, CA
- [Software Engineer, Fullstack](https://feeny.ai/job/software-engineer-fullstack-sciforium-san-francisco-w1txtdc4f5gz) — San Francisco, CA
- [GPU Cluster Engineer, Networking](https://feeny.ai/job/gpu-cluster-engineer-networking-sciforium-san-francisco-594atsvfkbpv) — San Francisco, CA
- [GPU Cluster Engineer, Systems & Platform](https://feeny.ai/job/gpu-cluster-engineer-systems-platform-sciforium-san-francisco-ma6a8rs3139z) — San Francisco, CA
- [GPU Cluster Engineer, Hardware Operations](https://feeny.ai/job/gpu-cluster-engineer-hardware-operations-sciforium-san-jose-pvp6n7qxehte) — San Jose, CA