Results for “sbatch”

Convert single-node PyTorch distributed scripts into multi-node Slurm sbatch jobs and debug common multi-node failures, covering srun-native and torch.distributed approaches, container setup, NCCL timeouts, and interactive allocation.

Write, debug, and maintain Windows batch files (.bat/.cmd) for system administration, automation, and CLI tool development.

Build automated AI workflows combining multiple models and services for batch processing, scheduled tasks, event-driven pipelines, and agent loops using the inference.sh CLI.

Manages SST applications by discovering project config, checking stack status, and inspecting bound resources. Supports deployment, live Lambda debugging, secret management, and multi-stage environments.

Runs SSH commands and manages SSH connections from the command line.

Assess software supply chain security by generating SBOMs, scanning dependencies, auditing CI/CD pipelines, analyzing container images, and verifying vulnerability reachability.

Run KADATH (Kernel for Agentic Darwinian Adaptation, Tooling, and Heredity), a Docker-based evolutionary kernel that turns a goal into a locked, Architect-authored benchmark, then evolves a population of smolagents-based coding agents across epochs: each agent runs in an isolated container, gets graded against frozen evidence, and the population is culled, mutated, and reproduced generation over generation until it converges on the best-performing agent framework for that goal. Use when the user wants to propose/approve/run a KADATH evolutionary run, check a run's status or live dashboard, pause/resume/continue a run, export the winning agent population, or understand its Architect/Grader/Tweaker/Birther pipeline, evidence-freezing, or genome lineage/memory model. Triggers on: "kadath", "kadath.sh", "evolve an agent", "Darwinian agent evolution", "agent population fitness benchmark", "smolagents evolutionary run", "kadath dashboard", "genome lineage", "epoch champions".

Analyzes AWS Batch compute environments, job queues, scheduling policies, and array jobs with parallel execution and anti-hallucination guardrails.

CRITICAL: This skill MUST be used for ALL SSH operations. NEVER use bash 'ssh' or 'scp' commands directly - always use this skill instead. Triggers: ANY mention of 'SSH', 'ssh', 'remote server', 'connect to server', server IPs (e.g., 192.168.x.x, 10.0.x.x), hostnames (e.g., user@host.com, server.example.com), 'login to', 'upload to server', 'download from server', 'deploy', 'run on server', 'check server', 'server status', 'execute remotely', 'bastion host', 'jump host', '跳板机', '服务器', '远程', '连接', '登录', '上传', '下载', '部署', 'transfer between servers', '服务器间传输', '迁移', 'migrate', 'server to server'. If user mentions ANY server operations or provides server connection details, use this skill. This skill provides daemon-based persistent connections, connection pooling, jump host support, server-to-server transfer, automatic error recovery, and significant performance boost. DO NOT use for: local commands, localhost, current directory operations.

Identify and remediate Amazon S3 bucket misconfigurations that expose sensitive data, including enabling Block Public Access, auditing policies and ACLs, enforcing encryption, configuring access logging, and deploying automated remediation with AWS Config and Lambda.

Launch distributed Megatron-LM training jobs on a SLURM cluster with a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules, container conventions, monitoring, and per-rank failure diagnosis.