Results for “research”

Autonomous discovery of concepts, methods, and vocabulary we have NOT heard of yet — sweeps broadly, diffs against a known-concepts ledger, and reports only what is genuinely new plus whether it names a gap in this stack. Built because 'graph engineering' had to arrive by word of mouth: a fixed topic list can only refresh what you already named, so it can never surface the thing you did not know to look for. Use for periodic/scheduled discovery, 'what's new in agent engineering', 'what are we missing', 'anything we haven't heard of'. Triggers: radar, discover, what's new, unknown unknowns, 新概念, 我们没听过的, 有什么没跟上的 — NOT for researching a topic you can already name (use /research), NOT for model releases (use /model-research), NOT for internal priorities (use /next).

Orchestrate token-heavy research, coding, and testing by delegating bounded tasks to cheaper subagents while reserving Claude Fable for architecture, synthesis, and final review.

Orchestrates building a new feature end-to-end by delegating research, planning, TDD implementation, review, and gated commit to specialized agents.

Plan a content strategy that drives traffic, builds authority, and generates leads by creating searchable or shareable content.

Query documented public database APIs with explicit endpoints, filters, pagination, and provenance for reproducible retrieval of scientific, regulatory, or financial facts.

Guides statistical hypothesis testing with assumption checks, effect sizes, power analysis, Bayesian alternatives, and APA-formatted reporting for research data.

Enables zero-shot image classification, image-text matching, and cross-modal retrieval using OpenAI's CLIP model.

Provides expert guidance for fast fine-tuning with Unsloth, including LoRA/QLoRA optimization, with 2-5x faster training and 50-80% less memory usage.

Train large-scale MoE models with FP8/INT4 low-precision RL, speculative decoding, and train-inference alignment using the miles framework.

Run ML training and batch jobs across multiple clouds with automatic cost optimization, spot instance recovery, and unified orchestration.

Generates publication-quality figures for ML papers, including architecture diagrams via Gemini and data-driven charts via matplotlib/seaborn.

Perform causal interventions on PyTorch models using pyvene's declarative framework for causal tracing, activation patching, and interchange intervention training.

Wraps the notebooklm-py CLI to ingest sources and generate synthesized artifacts like podcasts, slide decks, and quizzes from Google NotebookLM.

Identifies and prioritizes knowledge gaps, then systematically consults workspace files, past conversations, documentation, and web search to fill them with citable, confidence-rated answers.

Collects market data from Yahoo Finance and Exa, calculates technical indicators, scans watchlists, and generates market analysis reports to identify trading opportunities.

Delegates scientific literature searches to the Edison API, which autonomously searches hundreds of papers and returns a cited answer. Handles query submission, polling, and reporting results.

绘制研究领域或技术主题的谱系脉络与历史演进图，可视化思想的演化路径，展示早期工作中的技术难题如何被后续研究逐步解决。当用户想了解某个主题的发展轨迹、某个模型或技术的"家族树"（family tree）、某条研究线索在多年间的演进路线、技术迭代脉络、论文/模型谱系，或询问"X 是如何一步步发展来的""X 解决了前人的什么问题""梳理 X 的发展历史"时触发。产出为嵌入 Mermaid 图表的 Markdown 文件（演进图 + 节点明细表 + 阶段叙事），可直接粘贴至 Mermaid 查看器或提交至代码仓库。

Finding and expanding keyword opportunities for free — mining Search Console queries for high-impression/low-position pages, expanding via Google Autocomplete, and snapshotting SERPs with web search; optional paid DataForSEO volume/difficulty data.

Search YouTube SERP results through AISA for video research, channel discovery, trend checking, and competitor tracking. Use when: you need ranked YouTube results for a query, optionally filtered by country or language.

Multi-source search for agents across web, scholar, Tavily, and Perplexity Sonar endpoints. Use when you need structured retrieval, citation-backed answers, or broad research coverage from one AIsa API key.

Generate a comprehensive SEO strategy and implementation roadmap for new or existing websites, including competitive analysis, content planning, and technical foundation.

Access 31 pay-per-use tools for quantum security, agent safety, research, data, and earn on Base, paid via x402 protocol.

Automates hypothesis generation and testing on tabular datasets using LLMs, combining data-driven discovery with literature integration for scientific research.

Guide product managers through structured PRD creation by orchestrating problem framing, user research synthesis, solution definition, and success criteria into a cohesive document.

Store and query embeddings with metadata filtering, vector search, and full-text search using an open-source database that scales from notebooks to production.

Train large language models (7B-70B+) with RLHF using PPO, GRPO, DPO, and other algorithms, accelerated by Ray and vLLM for distributed multi-GPU setups.

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B.

Provides expert guidance for distributed training with DeepSpeed, covering ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, and sparse attention.

Add distributed training support to any PyTorch script with minimal code changes using a unified API for DDP, DeepSpeed, FSDP, and mixed precision.

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques for processing long documents and implementing efficient positional encodings.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques for 1.5-3.6× speedup without quality loss.

Design a usability test plan with tasks, success metrics, participant criteria, and facilitation guide for moderated or unmoderated sessions.

Guides authorized Active Directory security research covering Kerberos attacks, AD CS vulnerabilities, BloodHound path analysis, NTLM relay, and domain privilege escalation techniques.

Routes product-discovery work into the right stage — opportunity brief, research, PRD, tickets, or outcome review — and hands off to the matching sub-skill.

Audits and improves search visibility through technical SEO, on-page optimization, structured data, Core Web Vitals, and content strategy.

PdfLatex is a tool that converts Latex sources into PDF. This is specifically very important for researchers, as they use it to publish their findings. It could be installed very easily using Linux te