Find the perfect skill

Data & AI

Handles ingestion of RFP opportunities from sources such as SAM.gov, eMMA, and RFPMart, including CSV uploads via Admin UI. Normalizes all records to a canonical schema and deduplicates based on externalId and source. Use when adding new data sources, modifying ingestion logic, or debugging data fetching issues.

Data & AI

IDSTune is a reusable agent skill for integrated database tuning. It collects workload context, generates configuration, index, and materialized-view recommendations, and orchestrates optimization plans.

Data & AI

Checks mathematical correctness of TDA results using persistence diagram sanity, null model consistency, and benchmark comparisons.

Data & AI

Collects Douyin creator data (followers, video/image/live lists with likes/comments/shares) into a central Excel file and auto-refreshes monitoring dashboards.

Data & AI

Scaffold matplotlib figure code following trajectory_tda/viz conventions. Ensures consistent style, colors, sizes, and save patterns for publication figures.

Data & AI

Execute AI-generated code in secure isolated cloud sandboxes. Supports Python/JS/TS/R/Java/Bash, coding agents, Git, MCP Gateway, file operations, and sandbox lifecycle management.

Data & AI

Investigate failed or suspicious agent runs using production telemetry in D1. Query agent_runs and llm_call_metrics tables to find root causes.

Data & AI

Weekly LLM spend and unit-economics report, analyzing costs by stage, product, and household, to surface margin erosion before it becomes a loss.

Data & AI

Functional annotation and taxonomy inference from sequence homology.

Data & AI

Continuously monitors RL training logs via tail -f, extracting key metrics (reward, loss, episode length, success rate) and flagging anomalies like gradient explosion or convergence stalls. A research focus is scanning rollout and judge outputs to detect reward hacking patterns, systematic biases, and high-scoring cases mismatched with human preferences. Helps during exploratory RL experiments that require close inspection of undesirable model behaviors.

Data & AI

Profile ClickHouse queries to find CPU hotspots using system.trace_log and the sampling query profiler.

Data & AI

Create publication-quality charts and graphs for economics papers using clean styling, consistent scales, and export-ready formats.