polyaxon

AI control plane to schedule, run, debug, observe, and automate AI/ML workloads, agents, and sandboxes on your infra. Community:

[polyaxon @PolyaxonAI]

Programmatically :

pty = sandbox.pty.create(command=["sh"], cols=120, rows=40)
ws = sandbox.pty.attach(pty.pty_id, replay_bytes=0)

from polyaxon._pty.sandbox import SandboxPseudoTerminal

exit_code = SandboxPseudoTerminal(ws).start()

[polyaxon @PolyaxonAI]

Polyaxon's new sandbox client:

from polyaxon.client import SandboxClient

sandbox = SandboxClient(project="demo")
https://t.co/krwRMQVZJr()
result = sandbox.process.exec(["sh", "-lc", "echo ok"])
print(result.stdout)
print(result.stderr)
print(result.exit_code)

[polyaxon @PolyaxonAI]

And much simpler access with the CLI

polyaxon sandbox shell

[polyaxon @PolyaxonAI]

Fs ops:

sandbox.fs.write_text("/tmp/hello.txt", "hello\n")
https://t.co/45VSiljN6X_text("/tmp/hello.txt")
https://t.co/YZ4OSCC62y("/tmp")
sandbox.fs.upload_file("./local.txt", "/tmp/f.txt")
https://t.co/rvhlk5CTvO_file("/tmp/f.txt", "./copy.txt")

[polyaxon @PolyaxonAI]

Background exec:

bg = sandbox.process.exec_bg(["sh", "-lc", "while true; do date; sleep 1; done"], tag="clock")
print(https://t.co/IkSVXWaJUs)

for chunk in bg.iter_stdout(timeout=10):
print(chunk, end="")

bg.kill()

[polyaxon @PolyaxonAI]

Two new CLI commands are coming in beta in the next Polyaxon version:

`polyaxon sandbox` to launch isolated environments
`polyaxon ssh` to attach to workloads with governed remote access

Built for faster debugging loops across AI/ML workloads and agents.

[polyaxon @PolyaxonAI]

We’ve been quiet here. That changes today.

Polyaxon v3 is in active development: an AI control plane to schedule, run, debug, observe, and automate AI/ML workloads, agents, and sandboxes on your infrastructure.

We’ll share the work as it lands.

[polyaxon @PolyaxonAI]

How Polyaxon manages experiment tracking
https://t.co/zdSbEhyefs

[polyaxon @PolyaxonAI]

➡️ DevOps unifies software development and operations for faster, more reliable software delivery.
➡️ DataOps improves collaboration and automation for data analytics.
➡️ MLOps improves collaboration and automation for machine learning. #DevOps #DataOps #MLOps

[polyaxon @PolyaxonAI]

At Polyaxon we are building a set of open-source tools to help you manage and accelerate the complete life cycle of your deep learning and machine learning. https://t.co/u7ZTno342u

[polyaxon @PolyaxonAI]

As AI models become more prevalent in decision-making processes, it's important to consider the potential vulnerabilities that come with them. Without proper trust, risk, and security management, AI models are at risk of adversarial attacks. #MLOps #XAI #ModelMonitoring

[polyaxon @PolyaxonAI]

In short, AI model management without proper trust, risk, and security management is vulnerable to adversarial attacks. By prioritizing these areas, organizations can better protect their AI models and ensure their accuracy and reliability. #MLOps

[polyaxon @PolyaxonAI]

As the volume and complexity of our machine learning workloads continue to increase, it's crucial that the platform can handle this growth. The #MLOps platform needs to be scalable and flexible.

[polyaxon @PolyaxonAI]

Implementing MLOps in your organization using automation and integration of key processes can greatly reduce the time and the effort required to build and deploy machine learning models. #MLOps

[polyaxon @PolyaxonAI]

Monitoring and evaluation of machine learning models perfomance is important to know if they are addressing pain points as expected and to identify and fix any potential issues. #modelmonitoring