Staff Software Engineer, Infrastructure
Company: Anthropic Limited
Location: San Francisco
Posted on: November 6, 2024
Job Description:
About the role:
Anthropic is seeking talented and experienced Infrastructure
Engineers to join our team and support the development, scaling,
and maintenance of our cutting-edge AI systems. By joining our
Infrastructure team, you will have the opportunity to work on
groundbreaking AI technologies and contribute to the development of
frontier models, supporting Anthropic's mission to create safe and
reliable AI systems that benefit humanity.
We currently have openings on:
- Data Infrastructure: The Data Infrastructure team is
responsible for designing, building, and maintaining the data
infrastructure that powers our AI research and products. You will
collaborate with cross-functional teams to understand data
requirements, deliver efficient and reliable data solutions, and
continuously improve our data infrastructure. Your role will
involve building and optimizing data pipelines, implementing data
governance best practices, monitoring and troubleshooting, and
setting technical strategies for high-scale, reliable data
infrastructure and pipelines. You will work with technologies such
as Spark, Airflow, dbt, and cloud services from GCP and AWS, while
designing processes to ensure effective team operation and
continuous improvement.
- Research Infrastructure: The research infra team addresses the
problem of developing and scaling systems that enable researchers
to iterate quickly and also scale key systems/ components used by
researchers during the development phase to work at production
scale as our model footprint grows.
- Site Reliability Engineering: As an SRE at Anthropic, you will
design and implement scalable solutions, collaborate with
development teams to improve infrastructure reliability, and
establish monitoring systems, SLOs, and SLIs. You will implement
fault-tolerant design patterns, build automation tools, and
participate in an on-call rotation. Utilizing IaC principles, you
will collaborate with cross-functional teams to ensure reliability
and scalability in new features and services, and accelerate
engineering reliability through excellent tooling.
- Systems: The systems team is responsible for supporting some of
the largest, most sophisticated clusters in industry used to train,
research, and ultimately serve AI models. Your work will be crucial
in ensuring Anthropic is able to continue reliably and safely
training frontier models. You will be responsible for building
systems and running large Kubernetes clusters with GPU/TPU/Tranium
workloads.
- Observability: Observability team is responsible for designing,
building, and maintaining the observability infrastructure that
ensures the reliability, performance, and efficiency of our AI
systems and services. You will collaborate with cross-functional
teams to understand their observability requirements and deliver
solutions using technologies such as Prometheus, Splunk, Cloud
Logging, Grafana, and Honeycomb. Your role will involve developing
a config-driven approach to manage dashboards and alerts,
implementing structured logging and tracing, optimizing the
observability stack, and building a reliable system that requires
minimal maintenance. You will foster a culture of operational
excellence, proactive monitoring, and continuous improvement by
providing managed, centralized, and usable observability tools.
Responsibilities:
- Lead build out of industry-leading AI clusters (thousands to
hundreds of thousands of machines), partnering closely with cloud
service providers on cluster build out and required
features
- Consult with different stakeholders to deeply understand
infrastructure, data and compute needs, identifying potential
solutions to support frontier research and product
development
- Set technical strategy and oversee development of high scale,
reliable infrastructure systems.
- Mentor top technical talent
- Design processes (e.g. postmortem review, incident response,
on-call rotations) that help the team operate effectively and never
fail the same way twice
You may be a good fit if you:
- Have 10+ years of relevant industry experience, 3+ years
leading large scale, complex projects or teams as an engineer or
tech lead
- Are obsessed with distributed systems at scale, infrastructure
reliability, scalability, security, and continuous
improvement
- Strong proficiency in at least one programming language (e.g.,
Python, Rust, Go, Java)
- Strong problem-solving skills and ability to work
independently
- Have a passion for supporting internal partners like research
to understand their needs
- Have excellent communication skills to build consensus with
stakeholders, both internally and externally
- Possess deep knowledge of modern cloud infrastructure including
Kubernetes, Infrastructure as Code, AWS, and GCP
Strong candidates may also:
- Have security and privacy best practice expertise
- Experience with machine learning infrastructure like GPUs,
TPUs, or Trainium, as well as supporting networking infrastructure
like NCCL
- Low level systems experience, for example linux kernel tuning
and eBPF
- Technical expertise: Quickly understanding systems design
tradeoffs, keeping track of rapidly evolving software systems
Deadline to apply:None. Applications will be reviewed on a rolling
basis.
#J-18808-Ljbffr
Keywords: Anthropic Limited, Vallejo , Staff Software Engineer, Infrastructure, IT / Software / Systems , San Francisco, California
Didn't find what you're looking for? Search again!
Loading more jobs...