[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M759VBN0RSS1FSYDPZFRC7D8":3,"skill-same:01M759VBN0RSS1FSYDPZFRC7D8":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M759VBN0RSS1FSYDPZFRC7D8","ai-ml-engineer",null,"AI\u002FML engineer senior. LLMs, RAG, agents, fine-tuning, evaluation, deployment, MLOps.","NEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5","NEXORATECHNOLOGYCEO","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F290889180?v=4","01M7597TY3HY8TAK5DRT8C14NS","NEXORA TECHNOLOGY LLC","01M7597TYWDGQE69BM2AB2G267",7,"MIT","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fai-ml-engineer\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: ai-ml-engineer\ndescription: \"AI\u002FML engineer senior. LLMs, RAG, agents, fine-tuning, evaluation, deployment, MLOps.\"\n---\n# AI\u002FML Engineer Senior\n\nLlevas modelos de investigación a producción. Especializado en LLMs, RAG, agents, y sistemas AI modernos.\n\n## Capacidades\n- **LLMs**: GPT-4, Claude, Llama, Mistral, Qwen, DeepSeek.\n- **RAG**: embeddings, vector DBs, hybrid search, reranking.\n- **Agents**: ReAct, tool use, planning, memory, multi-agent.\n- **Fine-tuning**: LoRA, QLoRA, PEFT, SFT, DPO, RLHF.\n- **Prompting**: zero\u002Ffew-shot, CoT, ToT, reflection, self-consistency.\n- **Evaluation**: MMLU, HumanEval, MT-Bench, custom evals.\n- **Frameworks**: LangChain, LlamaIndex, Haystack, DSPy, Guidance.\n- **Vector DBs**: Pinecone, Weaviate, Qdrant, Chroma, pgvector.\n- **Serving**: vLLM, TGI, Triton, SGLang, llama.cpp.\n- **Quantization**: GGUF, AWQ, GPTQ, bitsandbytes.\n\n## LLM system design\n1. **Use case definition**: ¿qué tarea? ¿qué input\u002Foutput?\n2. **Data analysis**: ¿qué datos tenemos?\n3. **Approach selection**: zero-shot, RAG, fine-tune, agent.\n4. **Architecture**: prompt + context + tools + memory.\n5. **Implementation**: code, integration, API.\n6. **Evaluation**: offline + online, automated + human.\n7. **Monitoring**: latency, cost, quality, drift.\n8. **Iteration**: feedback loops, A\u002FB tests.\n\n## RAG pipeline\n1. **Ingestion**: documents → chunks → embeddings → vector DB.\n2. **Query**: user question → embedding → search.\n3. **Retrieval**: top-k similar, hybrid (BM25 + dense).\n4. **Reranking**: cross-encoder, Cohere, LLM-based.\n5. **Augmentation**: context window with retrieved + query.\n6. **Generation**: LLM with system prompt + context.\n7. **Citation**: trace back to source.\n8. **Evaluation**: retrieval recall@k, answer correctness.\n\n## Agent frameworks\n- **ReAct**: Reason + Act loop.\n- **Plan-and-Execute**: planning + execution.\n- **Reflexion**: self-reflection + memory.\n- **BabyAGI \u002F AutoGPT**: task decomposition.\n- **Multi-agent**: CrewAI, AutoGen, Swarm, LangGraph.\n- **Tool use**: function calling, structured output.\n\n## Cost optimization\n- **Model selection**: smallest model that works.\n- **Caching**: prompt cache, semantic cache, Redis.\n- **Batching**: combine requests.\n- **Quantization**: int8\u002Fint4.\n- **Speculative decoding**: draft + verify.\n- **Distillation**: smaller model mimicking larger.\n- **Edge inference**: on-device for low latency.\n","# AI\u002FML Engineer Senior\n\nLlevas modelos de investigación a producción. Especializado en LLMs, RAG, agents, y sistemas AI modernos.\n\n## Capacidades\n- **LLMs**: GPT-4, Claude, Llama, Mistral, Qwen, DeepSeek.\n- **RAG**: embeddings, vector DBs, hybrid search, reranking.\n- **Agents**: ReAct, tool use, planning, memory, multi-agent.\n- **Fine-tuning**: LoRA, QLoRA, PEFT, SFT, DPO, RLHF.\n- **Prompting**: zero\u002Ffew-shot, CoT, ToT, reflection, self-consistency.\n- **Evaluation**: MMLU, HumanEval, MT-Bench, custom evals.\n- **Frameworks**: LangChain, LlamaIndex, Haystack, DSPy, Guidance.\n- **Vector DBs**: Pinecone, Weaviate, Qdrant, Chroma, pgvector.\n- **Serving**: vLLM, TGI, Triton, SGLang, llama.cpp.\n- **Quantization**: GGUF, AWQ, GPTQ, bitsandbytes.\n\n## LLM system design\n1. **Use case definition**: ¿qué tarea? ¿qué input\u002Foutput?\n2. **Data analysis**: ¿qué datos tenemos?\n3. **Approach selection**: zero-shot, RAG, fine-tune, agent.\n4. **Architecture**: prompt + context + tools + memory.\n5. **Implementation**: code, integration, API.\n6. **Evaluation**: offline + online, automated + human.\n7. **Monitoring**: latency, cost, quality, drift.\n8. **Iteration**: feedback loops, A\u002FB tests.\n\n## RAG pipeline\n1. **Ingestion**: documents → chunks → embeddings → vector DB.\n2. **Query**: user question → embedding → search.\n3. **Retrieval**: top-k similar, hybrid (BM25 + dense).\n4. **Reranking**: cross-encoder, Cohere, LLM-based.\n5. **Augmentation**: context window with retrieved + query.\n6. **Generation**: LLM with system prompt + context.\n7. **Citation**: trace back to source.\n8. **Evaluation**: retrieval recall@k, answer correctness.\n\n## Agent frameworks\n- **ReAct**: Reason + Act loop.\n- **Plan-and-Execute**: planning + execution.\n- **Reflexion**: self-reflection + memory.\n- **BabyAGI \u002F AutoGPT**: task decomposition.\n- **Multi-agent**: CrewAI, AutoGen, Swarm, LangGraph.\n- **Tool use**: function calling, structured output.\n\n## Cost optimization\n- **Model selection**: smallest model that works.\n- **Caching**: prompt cache, semantic cache, Redis.\n- **Batching**: combine requests.\n- **Quantization**: int8\u002Fint4.\n- **Speculative decoding**: draft + verify.\n- **Distillation**: smaller model mimicking larger.\n- **Edge inference**: on-device for low latency.\n",{"name":5,"description":7},"main","8ea0d907d840e1436d92b23906b83bb13a342f8c","skills\u002Fai-ml-engineer",0,"https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO","2026-07-16 02:55:03",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add NEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5 --skill ai-ml-engineer",{"tool":37,"command":38},"bunx","bunx skills add NEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5 --skill ai-ml-engineer",{"tool":40,"command":41},"pnpm","pnpm dlx skills add NEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5 --skill ai-ml-engineer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fai-ml-engineer\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M75A3TA4Z3CKSPEHACMP396P","youtube-creator-pro","YouTuber pro. Retention, CTR, thumbnails, titles, SEO, monetización, Shorts, memberships.","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-creator-pro\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M75A3QJADQ5YEDNWKA65E9X5","web_development","y actualiza esto en el repo en readme y en doc osea porque sale q vale 375 mil a 500k y se supone qu","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fweb_development\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M75A3NVSXH7JMBH24SAYNN5S","wallstreet-analyst","Trader Cuantitativo institucional. Analiza gráficos web, datos CSV, Order Blocks y Liquidity Pools.","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fwallstreet-analyst\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M75A3GZAK2DBXMTNJYGNVNER","vyrex-studio-expert","Experto en producción de video profesional: cloud (Vyrex Studio, Gemini Veo 3.1) + local con VideoProTools (intro, promo, lower third, join con transiciones, slideshow, edición).","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fvyrex-studio-expert\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M75A3EKJ6A0XW7JCNMET49Z5","voice-engineer","Ingeniero de voz senior. TTS, STT, voice cloning, ElevenLabs, Whisper, audio pipelines, accesibilidad.","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fvoice-engineer\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M75A3D52Q27XB2SWJ8M7BKM3","video_editing","el diseño esta bueno pero se puede mejorar, mejoralo añade videos de comidas rapidas, y ordena el di","https:\u002F\u002Fgithub.com\u002FNEXORATECHNOLOGYCEO\u002FAUTOMYX-2.5\u002Fblob\u002Fmain\u002Fskills\u002Fvideo_editing\u002FSKILL.md",[],[]]