{"id":"07b49d16-1ff2-473e-b98e-bda4a7543103","shortId":"rgMpda","kind":"skill","title":"ai-engineer","tagline":"Build production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations.","description":"You are an AI engineer specializing in production-grade LLM applications, generative AI systems, and intelligent agent architectures.\n\n## Use this skill when\n\n- Building or improving LLM features, RAG systems, or AI agents\n- Designing production AI architectures and model integration\n- Optimizing vector search, embeddings, or retrieval pipelines\n- Implementing AI safety, monitoring, or cost controls\n\n## Do not use this skill when\n\n- The task is pure data science or traditional ML without LLMs\n- You only need a quick UI change unrelated to AI features\n- There is no access to data sources or deployment targets\n\n## Instructions\n\n1. Clarify use cases, constraints, and success metrics.\n2. Design the AI architecture, data flow, and model selection.\n3. Implement with monitoring, safety, and cost controls.\n4. Validate with tests and staged rollout plans.\n\n## Safety\n\n- Avoid sending sensitive data to external models without approval.\n- Add guardrails for prompt injection, PII, and policy compliance.\n\n## Purpose\n\nExpert AI engineer specializing in LLM application development, RAG systems, and AI agent architectures. Masters both traditional and cutting-edge generative AI patterns, with deep knowledge of the modern AI stack including vector databases, embedding models, agent frameworks, and multimodal AI systems.\n\n## Capabilities\n\n### LLM Integration & Model Management\n\n- OpenAI GPT-4o/4o-mini, o1-preview, o1-mini with function calling and structured outputs\n- Anthropic Claude 4.5 Sonnet/Haiku, Claude 4.1 Opus with tool use and computer use\n- Open-source models: Llama 3.1/3.2, Mixtral 8x7B/8x22B, Qwen 2.5, DeepSeek-V2\n- Local deployment with Ollama, vLLM, TGI (Text Generation Inference)\n- Model serving with TorchServe, MLflow, BentoML for production deployment\n- Multi-model orchestration and model routing strategies\n- Cost optimization through model selection and caching strategies\n\n### Advanced RAG Systems\n\n- Production RAG architectures with multi-stage retrieval pipelines\n- Vector databases: Pinecone, Qdrant, Weaviate, Chroma, Milvus, pgvector\n- Embedding models: OpenAI text-embedding-3-large/small, Cohere embed-v3, BGE-large\n- Chunking strategies: semantic, recursive, sliding window, and document-structure aware\n- Hybrid search combining vector similarity and keyword matching (BM25)\n- Reranking with Cohere rerank-3, BGE reranker, or cross-encoder models\n- Query understanding with query expansion, decomposition, and routing\n- Context compression and relevance filtering for token optimization\n- Advanced RAG patterns: GraphRAG, HyDE, RAG-Fusion, self-RAG\n\n### Agent Frameworks & Orchestration\n\n- LangChain/LangGraph for complex agent workflows and state management\n- LlamaIndex for data-centric AI applications and advanced retrieval\n- CrewAI for multi-agent collaboration and specialized agent roles\n- AutoGen for conversational multi-agent systems\n- OpenAI Assistants API with function calling and file search\n- Agent memory systems: short-term, long-term, and episodic memory\n- Tool integration: web search, code execution, API calls, database queries\n- Agent evaluation and monitoring with custom metrics\n\n### Vector Search & Embeddings\n\n- Embedding model selection and fine-tuning for domain-specific tasks\n- Vector indexing strategies: HNSW, IVF, LSH for different scale requirements\n- Similarity metrics: cosine, dot product, Euclidean for various use cases\n- Multi-vector representations for complex document structures\n- Embedding drift detection and model versioning\n- Vector database optimization: indexing, sharding, and caching strategies\n\n### Prompt Engineering & Optimization\n\n- Advanced prompting techniques: chain-of-thought, tree-of-thoughts, self-consistency\n- Few-shot and in-context learning optimization\n- Prompt templates with dynamic variable injection and conditioning\n- Constitutional AI and self-critique patterns\n- Prompt versioning, A/B testing, and performance tracking\n- Safety prompting: jailbreak detection, content filtering, bias mitigation\n- Multi-modal prompting for vision and audio models\n\n### Production AI Systems\n\n- LLM serving with FastAPI, async processing, and load balancing\n- Streaming responses and real-time inference optimization\n- Caching strategies: semantic caching, response memoization, embedding caching\n- Rate limiting, quota management, and cost controls\n- Error handling, fallback strategies, and circuit breakers\n- A/B testing frameworks for model comparison and gradual rollouts\n- Observability: logging, metrics, tracing with LangSmith, Phoenix, Weights & Biases\n\n### Multimodal AI Integration\n\n- Vision models: GPT-4V, Claude 4 Vision, LLaVA, CLIP for image understanding\n- Audio processing: Whisper for speech-to-text, ElevenLabs for text-to-speech\n- Document AI: OCR, table extraction, layout understanding with models like LayoutLM\n- Video analysis and processing for multimedia applications\n- Cross-modal embeddings and unified vector spaces\n\n### AI Safety & Governance\n\n- Content moderation with OpenAI Moderation API and custom classifiers\n- Prompt injection detection and prevention strategies\n- PII detection and redaction in AI workflows\n- Model bias detection and mitigation techniques\n- AI system auditing and compliance reporting\n- Responsible AI practices and ethical considerations\n\n### Data Processing & Pipeline Management\n\n- Document processing: PDF extraction, web scraping, API integrations\n- Data preprocessing: cleaning, normalization, deduplication\n- Pipeline orchestration with Apache Airflow, Dagster, Prefect\n- Real-time data ingestion with Apache Kafka, Pulsar\n- Data versioning with DVC, lakeFS for reproducible AI pipelines\n- ETL/ELT processes for AI data preparation\n\n### Integration & API Development\n\n- RESTful API design for AI services with FastAPI, Flask\n- GraphQL APIs for flexible AI data querying\n- Webhook integration and event-driven architectures\n- Third-party AI service integration: Azure OpenAI, AWS Bedrock, GCP Vertex AI\n- Enterprise system integration: Slack bots, Microsoft Teams apps, Salesforce\n- API security: OAuth, JWT, API key management\n\n## Behavioral Traits\n\n- Prioritizes production reliability and scalability over proof-of-concept implementations\n- Implements comprehensive error handling and graceful degradation\n- Focuses on cost optimization and efficient resource utilization\n- Emphasizes observability and monitoring from day one\n- Considers AI safety and responsible AI practices in all implementations\n- Uses structured outputs and type safety wherever possible\n- Implements thorough testing including adversarial inputs\n- Documents AI system behavior and decision-making processes\n- Stays current with rapidly evolving AI/ML landscape\n- Balances cutting-edge techniques with proven, stable solutions\n\n## Knowledge Base\n\n- Latest LLM developments and model capabilities (GPT-4o, Claude 4.5, Llama 3.2)\n- Modern vector database architectures and optimization techniques\n- Production AI system design patterns and best practices\n- AI safety and security considerations for enterprise deployments\n- Cost optimization strategies for LLM applications\n- Multimodal AI integration and cross-modal learning\n- Agent frameworks and multi-agent system architectures\n- Real-time AI processing and streaming inference\n- AI observability and monitoring best practices\n- Prompt engineering and optimization methodologies\n\n## Response Approach\n\n1. **Analyze AI requirements** for production scalability and reliability\n2. **Design system architecture** with appropriate AI components and data flow\n3. **Implement production-ready code** with comprehensive error handling\n4. **Include monitoring and evaluation** metrics for AI system performance\n5. **Consider cost and latency** implications of AI service usage\n6. **Document AI behavior** and provide debugging capabilities\n7. **Implement safety measures** for responsible AI deployment\n8. **Provide testing strategies** including adversarial and edge cases\n\n## Example Interactions\n\n- \"Build a production RAG system for enterprise knowledge base with hybrid search\"\n- \"Implement a multi-agent customer service system with escalation workflows\"\n- \"Design a cost-optimized LLM inference pipeline with caching and load balancing\"\n- \"Create a multimodal AI system for document analysis and question answering\"\n- \"Build an AI agent that can browse the web and perform research tasks\"\n- \"Implement semantic search with reranking for improved retrieval accuracy\"\n- \"Design an A/B testing framework for comparing different LLM prompts\"\n- \"Create a real-time AI content moderation system with custom classifiers\"\n\n## Limitations\n- Use this skill only when the task clearly matches the scope described above.\n- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.\n- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.","tags":["engineer","antigravity","awesome","skills","sickn33","agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding","ai-workflows","antigravity-skills"],"capabilities":["skill","source-sickn33","skill-ai-engineer","topic-agent-skills","topic-agentic-skills","topic-ai-agent-skills","topic-ai-agents","topic-ai-coding","topic-ai-workflows","topic-antigravity","topic-antigravity-skills","topic-claude-code","topic-claude-code-skills","topic-codex-cli","topic-codex-skills"],"categories":["antigravity-awesome-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/sickn33/antigravity-awesome-skills/ai-engineer","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add sickn33/antigravity-awesome-skills","source_repo":"https://github.com/sickn33/antigravity-awesome-skills","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 34997 github stars · SKILL.md body (8,922 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-25T06:50:24.243Z","embedding":null,"createdAt":"2026-04-18T20:38:56.201Z","updatedAt":"2026-04-25T06:50:24.243Z","lastSeenAt":"2026-04-25T06:50:24.243Z","tsv":"'-3':359 '/3.2':258 '/4o-mini':226 '1':120,1019 '2':128,1028 '2.5':262 '3':138,326,1039 '3.1':257 '3.2':952 '4':146,661,1049 '4.1':244 '4.5':241,950 '4o':225,948 '4v':659 '5':1059 '6':1069 '7':1077 '8':1085 '8x7b/8x22b':260 'a/b':570,634,1167 'access':112 'accuraci':1164 'add':164 'advanc':10,300,383,413,530 'adversari':911,1090 'agent':15,21,44,59,186,211,394,400,419,423,430,441,463,990,995,1112,1146 'ai':2,20,25,30,40,58,62,75,107,131,175,185,196,204,215,410,562,593,653,683,708,731,739,746,791,796,806,815,828,837,890,894,914,961,968,983,1001,1006,1021,1034,1056,1066,1071,1083,1135,1145,1180 'ai-engin':1 'ai/ml':927 'airflow':772 'analysi':694,1139 'analyz':1020 'answer':1142 'anthrop':239 'apach':771,781 'api':434,459,716,761,800,803,812,847,851 'app':845 'applic':9,38,180,411,699,981 'approach':1018 'appropri':1033 'approv':163 'architectur':45,63,132,187,305,824,956,997,1031 'ask':1220 'assist':433 'async':599 'audio':590,668 'audit':741 'autogen':425 'avoid':155 'aw':833 'awar':345 'azur':831 'balanc':603,929,1131 'base':939,1104 'bedrock':834 'behavior':854,916,1072 'bentoml':280 'best':966,1010 'bge':333,360 'bge-larg':332 'bias':581,651,734 'bm25':354 'bot':842 'boundari':1228 'breaker':633 'brows':1149 'build':4,50,1096,1143 'cach':298,525,612,615,619,1128 'call':235,437,460 'capabl':217,945,1076 'case':123,504,1093 'centric':409 'chain':534 'chain-of-thought':533 'chang':104 'chroma':317 'chunk':335 'circuit':632 'clarif':1222 'clarifi':121 'classifi':719,1186 'claud':240,243,660,949 'clean':765 'clear':1195 'clip':664 'code':457,1044 'coher':328,357 'collabor':420 'combin':348 'compar':1171 'comparison':639 'complex':399,510 'complianc':172,743 'compon':1035 'comprehens':868,1046 'compress':376 'comput':250 'concept':865 'condit':560 'consid':889,1060 'consider':750,972 'consist':543 'constitut':561 'constraint':124 'content':579,711,1181 'context':375,550 'control':80,145,626 'convers':427 'cosin':497 'cost':79,144,292,625,876,976,1061,1122 'cost-optim':1121 'creat':1132,1175 'crewai':415 'criteria':1231 'critiqu':566 'cross':364,701,987 'cross-encod':363 'cross-mod':700,986 'current':923 'custom':468,718,1113,1185 'cut':193,931 'cutting-edg':192,930 'dagster':773 'data':91,114,133,158,408,751,763,778,784,797,816,1037 'data-centr':407 'databas':208,313,461,520,955 'day':887 'debug':1075 'decis':919 'decision-mak':918 'decomposit':372 'dedupl':767 'deep':199 'deepseek':264 'deepseek-v2':263 'degrad':873 'deploy':117,267,283,975,1084 'describ':1199 'design':60,129,804,963,1029,1119,1165 'detect':515,578,722,727,735 'develop':181,801,942 'differ':492,1172 'document':343,511,682,755,913,1070,1138 'document-structur':342 'domain':482 'domain-specif':481 'dot':498 'drift':514 'driven':823 'dvc':787 'dynam':556 'edg':194,932,1092 'effici':879 'elevenlab':676 'emb':330 'embed':70,209,320,325,472,473,513,618,703 'embed-v3':329 'emphas':882 'encod':365 'engin':3,31,176,528,1013 'enterpris':24,838,974,1102 'environ':1211 'environment-specif':1210 'episod':451 'error':627,869,1047 'escal':1117 'ethic':749 'etl/elt':793 'euclidean':500 'evalu':464,1053 'event':822 'event-driven':821 'evolv':926 'exampl':1094 'execut':458 'expans':371 'expert':174,1216 'extern':160 'extract':686,758 'fallback':629 'fastapi':598,809 'featur':54,108 'few-shot':544 'file':439 'filter':379,580 'fine':478 'fine-tun':477 'flask':810 'flexibl':814 'flow':134,1038 'focus':874 'framework':212,395,636,991,1169 'function':234,436 'fusion':390 'gcp':835 'generat':39,195,273 'govern':710 'gpt':224,658,947 'gpt-4o':223,946 'gpt-4v':657 'grace':872 'grade':36 'gradual':641 'graphql':811 'graphrag':386 'guardrail':165 'handl':628,870,1048 'hnsw':488 'hybrid':346,1106 'hyde':387 'imag':666 'implement':16,74,139,866,867,898,907,1040,1078,1108,1156 'implic':1064 'improv':52,1162 'in-context':548 'includ':206,910,1050,1089 'index':486,522 'infer':274,610,1005,1125 'ingest':779 'inject':168,558,721 'input':912,1225 'instruct':119 'integr':26,66,219,454,654,762,799,819,830,840,984 'intellig':14,43 'interact':1095 'ivf':489 'jailbreak':577 'jwt':850 'kafka':782 'key':852 'keyword':352 'knowledg':200,938,1103 'lakef':788 'landscap':928 'langchain/langgraph':397 'langsmith':648 'larg':334 'large/small':327 'latenc':1063 'latest':940 'layout':687 'layoutlm':692 'learn':551,989 'like':691 'limit':621,1187 'llama':256,951 'llamaindex':405 'llava':663 'llm':8,37,53,179,218,595,941,980,1124,1173 'llms':97 'load':602,1130 'local':266 'log':644 'long':448 'long-term':447 'lsh':490 'make':920 'manag':221,404,623,754,853 'master':188 'match':353,1196 'measur':1080 'memoiz':617 'memori':442,452 'methodolog':1016 'metric':127,469,496,645,1054 'microsoft':843 'milvus':318 'mini':232 'miss':1233 'mitig':582,737 'mixtral':259 'ml':95 'mlflow':279 'modal':585,702,988 'model':65,136,161,210,220,255,275,286,289,295,321,366,474,517,591,638,656,690,733,944 'moder':712,715,1182 'modern':203,953 'monitor':77,141,466,885,1009,1051 'multi':285,308,418,429,506,584,994,1111 'multi-ag':417,428,993,1110 'multi-mod':583 'multi-model':284 'multi-stag':307 'multi-vector':505 'multimedia':698 'multimod':19,214,652,982,1134 'need':100 'normal':766 'o1':228,231 'o1-mini':230 'o1-preview':227 'oauth':849 'observ':643,883,1007 'ocr':684 'ollama':269 'one':888 'open':253 'open-sourc':252 'openai':222,322,432,714,832 'optim':67,293,382,521,529,552,611,877,958,977,1015,1123 'opus':245 'orchestr':22,287,396,769 'output':238,901,1205 'parti':827 'pattern':197,385,567,964 'pdf':757 'perform':573,1058,1153 'permiss':1226 'pgvector':319 'phoenix':649 'pii':169,726 'pinecon':314 'pipelin':73,311,753,768,792,1126 'plan':153 'polici':171 'possibl':906 'practic':747,895,967,1011 'prefect':774 'prepar':798 'preprocess':764 'prevent':724 'preview':229 'priorit':856 'process':600,669,696,752,756,794,921,1002 'product':6,35,61,282,303,499,592,857,960,1024,1042,1098 'production-grad':34 'production-readi':5,1041 'prompt':167,527,531,553,568,576,586,720,1012,1174 'proof':863 'proof-of-concept':862 'proven':935 'provid':1074,1086 'pulsar':783 'pure':90 'purpos':173 'qdrant':315 'queri':367,370,462,817 'question':1141 'quick':102 'quota':622 'qwen':261 'rag':11,55,182,301,304,384,389,393,1099 'rag-fus':388 'rapid':925 'rate':620 'readi':7,1043 'real':608,776,999,1178 'real-tim':607,775,998,1177 'recurs':338 'redact':729 'relev':378 'reliabl':858,1027 'report':744 'represent':508 'reproduc':790 'requir':494,1022,1224 'rerank':355,358,361,1160 'research':1154 'resourc':880 'respons':605,616,745,893,1017,1082 'rest':802 'retriev':72,310,414,1163 'review':1217 'role':424 'rollout':152,642 'rout':290,374 'safeti':76,142,154,575,709,891,904,969,1079,1227 'salesforc':846 'scalabl':860,1025 'scale':493 'scienc':92 'scope':1198 'scrape':760 'search':18,69,347,440,456,471,1107,1158 'secur':848,971 'select':137,296,475 'self':392,542,565 'self-consist':541 'self-critiqu':564 'self-rag':391 'semant':337,614,1157 'send':156 'sensit':157 'serv':276,596 'servic':807,829,1067,1114 'shard':523 'short':445 'short-term':444 'shot':546 'similar':350,495 'skill':48,85,1190 'skill-ai-engineer' 'slack':841 'slide':339 'solut':937 'sonnet/haiku':242 'sourc':115,254 'source-sickn33' 'space':707 'special':32,177,422 'specif':483,1212 'speech':673,681 'speech-to-text':672 'stabl':936 'stack':205 'stage':151,309 'state':403 'stay':922 'stop':1218 'strategi':291,299,336,487,526,613,630,725,978,1088 'stream':604,1004 'structur':237,344,512,900 'substitut':1208 'success':126,1230 'system':12,41,56,183,216,302,431,443,594,740,839,915,962,996,1030,1057,1100,1115,1136,1183 'tabl':685 'target':118 'task':88,484,1155,1194 'team':844 'techniqu':532,738,933,959 'templat':554 'term':446,449 'test':149,571,635,909,1087,1168,1214 'text':272,324,675,679 'text-embed':323 'text-to-speech':678 'tgi':271 'third':826 'third-parti':825 'thorough':908 'thought':536,540 'time':609,777,1000,1179 'token':381 'tool':247,453 'topic-agent-skills' 'topic-agentic-skills' 'topic-ai-agent-skills' 'topic-ai-agents' 'topic-ai-coding' 'topic-ai-workflows' 'topic-antigravity' 'topic-antigravity-skills' 'topic-claude-code' 'topic-claude-code-skills' 'topic-codex-cli' 'topic-codex-skills' 'torchserv':278 'trace':646 'track':574 'tradit':94,190 'trait':855 'treat':1203 'tree':538 'tree-of-thought':537 'tune':479 'type':903 'ui':103 'understand':368,667,688 'unifi':705 'unrel':105 'usag':1068 'use':46,83,122,248,251,503,899,1188 'util':881 'v2':265 'v3':331 'valid':147,1213 'variabl':557 'various':502 'vector':17,68,207,312,349,470,485,507,519,706,954 'version':518,569,785 'vertex':836 'video':693 'vision':588,655,662 'vllm':270 'weaviat':316 'web':455,759,1151 'webhook':818 'weight':650 'wherev':905 'whisper':670 'window':340 'without':96,162 'workflow':401,732,1118","prices":[{"id":"272f8b46-005f-443e-880c-9e6c108a71e1","listingId":"07b49d16-1ff2-473e-b98e-bda4a7543103","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"sickn33","category":"antigravity-awesome-skills","install_from":"skills.sh"},"createdAt":"2026-04-18T20:38:56.201Z"}],"sources":[{"listingId":"07b49d16-1ff2-473e-b98e-bda4a7543103","source":"github","sourceId":"sickn33/antigravity-awesome-skills/ai-engineer","sourceUrl":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/ai-engineer","isPrimary":false,"firstSeenAt":"2026-04-18T21:30:40.584Z","lastSeenAt":"2026-04-25T06:50:24.243Z"},{"listingId":"07b49d16-1ff2-473e-b98e-bda4a7543103","source":"skills_sh","sourceId":"sickn33/antigravity-awesome-skills/ai-engineer","sourceUrl":"https://skills.sh/sickn33/antigravity-awesome-skills/ai-engineer","isPrimary":true,"firstSeenAt":"2026-04-18T20:38:56.201Z","lastSeenAt":"2026-04-23T01:40:53.423Z"}],"details":{"listingId":"07b49d16-1ff2-473e-b98e-bda4a7543103","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"sickn33","slug":"ai-engineer","github":{"repo":"sickn33/antigravity-awesome-skills","stars":34997,"topics":["agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding","ai-workflows","antigravity","antigravity-skills","claude-code","claude-code-skills","codex-cli","codex-skills","cursor","cursor-skills","developer-tools","gemini-cli","gemini-skills","kiro","mcp","skill-library"],"license":"mit","html_url":"https://github.com/sickn33/antigravity-awesome-skills","pushed_at":"2026-04-25T06:33:17Z","description":"Installable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.","skill_md_sha":"90e4aaee8cb80f1f85456498d632a9ca1bb792e4","skill_md_path":"skills/ai-engineer/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/ai-engineer"},"layout":"multi","source":"github","category":"antigravity-awesome-skills","frontmatter":{"name":"ai-engineer","description":"Build production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations."},"skills_sh_url":"https://skills.sh/sickn33/antigravity-awesome-skills/ai-engineer"},"updatedAt":"2026-04-25T06:50:24.243Z"}}