Agent skills for DeepSeek Harness

Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow…

Design or audit a model-agnostic evaluation harness for an LLM or multimodal LLM on a clinical task (radiology report generation, visual…

When to activate: MLOps, DVC, model versioning, drift detection, Evidently, WhyLogs, retraining, ML CI/CD, model monitoring

Use when designing or auditing the evaluation of an MLSys paper, selecting representative workloads and hardware, tuning baselines…

Use when positioning an MLSys submission against the fast-moving ML-systems literature scattered across OSDI, SOSP, NSDI, ASPLOS, and ML…

Use when hardening the reproducibility of MLSys performance claims, pinning the full system layer from driver to interconnect, separating…

Use when reasoning about how MLSys peer review works, covering the OpenReview workflow, the mixed ML-and-systems reviewer pool and how…

Use when auditing an MLSys submission for OpenReview readiness, covering the two-column 10-page body excluding references, the separate…

Use when deciding what goes into an MLSys appendix versus the 10-page body, exploiting the venue's unlimited separately-uploaded appendix…

Use when revising an MLSys paper's prose and structure, building the measured-bottleneck opening, naming the mechanism instead of listing…

Kotlin 2.0+ development specialist covering Ktor, coroutines, Compose Multiplatform, and Kotlin-idiomatic patterns.

Firebase Firestore specialist covering NoSQL patterns, real-time sync, offline caching, and Security Rules.

Neon serverless PostgreSQL specialist covering auto-scaling, database branching, PITR, and connection pooling.

Supabase specialist covering PostgreSQL 16, pgvector, RLS, real-time subscriptions, and Edge Functions.

Use when packaging a MobiCom artifact for the evaluation committee — choosing among the three ACM badges as a calibration of what you can…

Mobile app development guide — React Native, Flutter, Expo for cross-platform.

Use cryptography correctly in a mobile app against OWASP Mobile Top 10 M10.

Harden how a mobile app stores data at rest so a lost, stolen, backed-up, or rooted/jailbroken device does not leak credentials, tokens…

Abuse insecure deep links and inter-app entry points during an authorized mobile assessment.

Streamlines the development, testing, and automation of mobile applications through specialized agentic workflows.

End-to-end mobile testing of Expo/React Native apps via claude-in-mobile MCP + mcporter.