Software Engineer, Fireworks AI

AI Training Infrastructure

$175-190k

AWS

Docker

Kubernetes

GCP

Azure

PyTorch

Mid and Senior level

San Francisco Bay Area

More information about location

3 days a week in office (Redwood City, CA)

Generative AI platform enabling rapid scaling for developers and businesses

Open for applications

Generative AI platform enabling rapid scaling for developers and businesses

21-100 employees

B2BArtificial IntelligenceSaaS

Open for applications

$175-190k

AWS

Docker

Kubernetes

GCP

Azure

PyTorch

Mid and Senior level

San Francisco Bay Area

More information about location

3 days a week in office (Redwood City, CA)

21-100 employees

B2BArtificial IntelligenceSaaS

Company mission

To help every organization unleash the full power of artificial intelligence

Job

Company

Role

Who you are

Bachelor's degree in Computer Science, Computer Engineering, or related field, or equivalent practical experience
3+ years of experience with distributed systems and ML infrastructure
Experience with PyTorch
Proficiency in cloud platforms (AWS, GCP, Azure)
Experience with containerization, orchestration (Kubernetes, Docker)
Knowledge of distributed training techniques (data parallelism, model parallelism, FSDP)

Desirable

Master's or PhD in Computer Science or related field
Experience training large language models or multimodal AI systems
Experience with ML workflow orchestration tools
Background in optimizing high-performance distributed computing systems
Familiarity with ML DevOps practices
Contributions to open-source ML infrastructure or related projects

What the job involves

As a Training Infrastructure Engineer, you'll design, build, and optimize the infrastructure that powers our large-scale model training operations. Your work will be essential to developing high-performance AI training infrastructure. You'll collaborate with AI researchers and engineers to create robust training pipelines, optimize distributed training workloads, and ensure reliable model development
Design and implement scalable infrastructure for large-scale model training workloads
Develop and maintain distributed training pipelines for LLMs and multimodal models
Optimize training performance across multiple GPUs, nodes, and data centers
Implement monitoring, logging, and debugging tools for training operations
Architect and maintain data storage solutions for large-scale training datasets
Automate infrastructure provisioning, scaling, and orchestration for model training
Collaborate with researchers to implement and optimize training methodologies
Analyze and improve efficiency, scalability, and cost-effectiveness of training systems
Troubleshoot complex performance issues in distributed training environments

Our take

Whilst the development of generative AI is thought to be the reserve of large companies like Open AI, smaller startups such as Firework.AI are making this technology easier to access and develop for smaller companies wanting to handle specific business needs using AI. As an open-source model API, it has one of the largest user bases - with over 12,000 users leveraging Fireworks.AI’s software.

Though lacking the brand name recognition of generative AI developers like Google and Meta, Fireworks.AI differentiates itself in a market crowded with large competitors through its unique approach to deploying its software. It creates smaller, production-grade, customisable generative AI models that can be launched privately and securely, rather than the non-private and inflexible models built by larger companies.

After its 2024 Series B round of $52 million, Fireworks.AI has been valued at over $550 million, an impressive sum for a David in a sea of Goliaths. The company plans to use its funding to develop compound AI systems that can orchestrate across multiple single models. It will be interesting to see how the company evolves with its potentially disruptive business approach to generative AI.

Freddie

Company Specialist at Welcome to the Jungle

Company

Funding (2 rounds)

Jul 2024

$52m

SERIES B

Mar 2024

$25m

SERIES A

Total funding: $77m

Company HQ

Redwood City, CA

Leadership

Lin Qiao

(CEO)

Previously Senior Director of Engineering at Meta, Tech Lead at Linkedin, Research Staff Member at IBM Almaden Research Center and an Advisory Software Engineer at IBM Silicon Valley Lab.

Salary benchmarks

We don't have enough data yet to provide salary benchmarks for this role.

Submit your salary to help other candidates with crowdsourced salary estimates.

Share this job

View 11 more jobs at Fireworks AI