Results for “dep”

Machine learning development agent for end-to-end ML workflows: data preprocessing, model training, evaluation, hyperparameter tuning, and deployment

Shortcut for security review on current change set. Runs layered checks (secret-scan, dep-audit, semgrep, OWASP patterns, prompt-injection review).

Computer vision engineering skill for object detection, image segmentation, and visual AI systems. Covers CNN and Vision Transformer architectures, YOLO/Faster R-CNN/DETR detection, Mask R-CNN/SAM segmentation, and production deployment with ONNX/TensorRT. Includes PyTorch, torchvision, Ultralytics, Detectron2, and MMDetection frameworks. Use when building detection pipelines, training custom models, optimizing inference, or deploying vision systems.

Nest.js framework expert specializing in module architecture, dependency injection, middleware, guards, interceptors, testing with Jest/Supertest, TypeORM/Mongoose integration, and Passport.js authentication. Use PROACTIVELY for any Nest.js application issues including architecture decisions, testing strategies, performance optimization, or debugging complex dependency injection problems. If a specialized expert is a better fit, I will recommend switching and stop.

Deploys and optimizes AI/ML inference workloads on GKE, using GPUs, TPUs, and model servers.

`analysis-agent`/`review-agent`: use when module ownership, public surface, dependency direction, cycles, or shared state changes; skip owner-internal placement.

`analysis-agent`/`review-agent`: use when proven variation needs extension points, plugins, policy hooks, compatibility, or deprecation; skip when no extension decision exists.

Deploy a cuOpt REST server and solve routing optimization problems using Python or curl.

Build and deploy machine learning models with TensorFlow, covering Keras, data pipelines, and production serving.

`analysis-agent`/`task-agent`/`review-agent`: use when API, schema, event, or behavior changes need compatibility, deprecation, rollout, or migration; skip without version impact.

`analysis-agent`/`task-agent`/`review-agent`: use when a service split affects ownership, deployment, scaling, isolation, contracts, or data; skip without a split decision.

Audit a proposed assessment for construct validity, reliability, and alignment to learning objectives. Use when reviewing or quality-assuring assessments before deployment.

Guides attorneys through deposition witness preparation using a two-session model with document review, practice examination, and day-of logistics. Covers party witnesses, fact witnesses, 30(b)(6) corporate representatives, and experts. Produces preparation memos, document review lists, topic summaries, and day-of checklists. Enforces ABA Opinion 508 ethical boundaries. Use when preparing any witness for deposition, scheduling prep sessions, or building witness preparation materials.

Instrument LLM applications with OpenInference tracing for Phoenix AI observability, covering setup, custom spans, and production deployment.

Runs ComfyUI image-generation workflows from an AI agent via a CLI, handling imports, dependencies, execution, and history.

`task-agent`/`review-agent`: use when protected state, restore objectives, dependency order, or recovery evidence changes; skip backup-job-only work with no recovery decision.

MLOps Engineer IA — Expert en déploiement de modèles ML (A/B testing, feature stores, MLflow, Kubeflow). Industrialise le ML du notebook à la production.

Operate long-lived agent workloads with observability, security boundaries, and lifecycle management.

Automate BigML machine learning operations through Composio's toolkit via Rube MCP, including model creation, training, and deployment.

`analysis-agent`/`task-agent`/`review-agent`: use when migration, backfill, deployment order, rollback, or live-data safety needs design; skip when no data migration exists.

Select, validate, patch, and deploy existing NVIDIA Dynamo Kubernetes recipes for model serving with GPU support.

Build, deploy, and manage continuous AI agents using a visual workflow editor or development toolkit.

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency on NVIDIA GPUs (A100/H100).

Self-hosted web crawling and content extraction exposed as MCP tools, with depth control and clean markdown output.

Orchestrate teams of agents — DAG execution, routing, state sharing, failure recovery. Use when coordinating parallel agents in task pipelines with dependencies or distributing work by capability.

Plan and execute large refactor or rewrite efforts efficiently with parallel multi-agent analysis and implementation.

Explains machine learning model predictions using SHAP values, covering feature importance, visualizations, debugging, bias analysis, and production deployment.

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

Generate scaffolds that gradually increase student choice, voice, and ownership within a learning task. Use when students depend heavily on teacher direction and need to develop autonomy.

Run config-driven performance benchmarks against a deployed NVIDIA RAG Blueprint server, including profiling and load testing, with a unified report.

Train, fine-tune, and deploy LLMs using LitGPT's clean implementations of 20+ architectures like Llama, Gemma, and Phi.

Deploy Meta's LlamaGuard moderation model to filter LLM inputs and outputs across 6 safety categories using HuggingFace, vLLM, or FastAPI.

Refactor legacy codebases, migrate outdated frameworks, and implement gradual modernization. Handles technical debt, dependency updates, and backward compatibility. Use PROACTIVELY for legacy system updates, framework migrations, or technical debt reduction.

Seven-role depersonalised panel review of framework artefacts (KUD, criterion bank, LT definition, crosswalk, scope-and-sequence) in sequential-isolation mode. Gate rule mean>=88 AND no role<70.

Manages Agent Platform serving endpoints: create, list, describe, update, delete, and troubleshoot permission, quota, and resource busy errors.