Tips menjalani derasnya perkembangan AI, dari pengalaman gw sendiri:
https://t.co/v8NBHg0uX0
1/ Gak perlu ngikutin semua update yg keluar. Gw sendiri cuma fokus pake beberapa tools aja, kayak Claude & Cursor. Itu udah cukup buat kerjaan harian.
2/ Ada banyak trend yg lewat tanpa sempet gw coba, misal OpenClaw yg dari hype sampe redup, gw cuma nonton dari pinggir. Gak masalah, gak semua hype harus dikejar. Gw ujungnya pake Claude Desktop/Dispatch juga.
3/ Timebox waktu belajar. Misal 5 jam per minggu khusus buat explore tools atau teknik AI baru, jangan sampai overflow ke waktu kerja atau istirahat.
4/ Dari 5 jam itu, identify dulu apakah tool baru ini bakal ngebantu kerjaan lo atau cuma sekedar "nice to know". Kalo emang relevan & bisa langsung dipake, prioritasin (misal 3 dari 5 jam). Kalo cuma buat tau aja, kasih porsi kecil (2 jam).
5/ Setelah dipake, ukur ROI atau seberapa cepat kerjaan jadi lebih efisien. Kalo gak ningkatin output, ya gausah dipaksain dipake terus.
6/ Jangan terlalu cepat gonta ganti tools. Stick with one for quite a while, karena mungkin lo emang belum terbiasa aja, bukan tools-nya yg jelek. Switching terus malah bikin lo gak pernah beneran dalem di satu tools.
Intinya, lo gak harus jadi orang pertama yg coba semua tools baru. Yg penting konsisten upgrade skill dgn cara yg measurable & gak bikin burnout.
Seorang siswa gagal menyelesaikan kompetisi OSN karena terjadi pemadaman listrik total di menit ke32 ketika pengerjaan soal.
Pemadaman berlangsung lebih dari 4 jam & memutus seluruh jaringan internet di beberapa kecamatan, sehingga data seluler pun tidak bisa digunakan untuk mengerjakan soal.
CC: nanad.0814
Naive RAG vs. Agentic RAG, explained visually:
Naive RAG breaks in 3 ways:
↳ It retrieves once and generates once. If the context isn't relevant, the system can't search again.
↳ It treats every query the same. A simple lookup and a multi-hop reasoning task go through the identical retrieve-then-generate path.
↳ And there's no verification. Whatever the retriever returns gets blindly trusted.
Agentic RAG fixes this by introducing decision-making loops at each stage.
Steps 1-2) A query rewriting agent reformulates the raw query. This goes beyond fixing typos. It makes vague terms precise, decomposes complex queries into sub-queries, and expands abbreviations.
Steps 3-5) A routing agent decides if the query even needs external context. If not, retrieval is skipped. If yes, a source selector picks the best backend for this specific query type.
Steps 6-7) The source selector routes to the most appropriate source. Vector DB for semantic search, web search for real-time info, or structured APIs for tabular data. The retrieved context and rewritten query are combined into the prompt.
Steps 8-9) The LLM generates an initial response.
Steps 10-12) A validation agent (Corrective RAG) checks whether the response is relevant, grounded, and complete. If it passes, it's returned. If not, the system loops back to Step 1 with a reformulated query.
This continues for some iterations until we get a satisfactory response or the system admits it cannot answer.
The reason it works is that each agent acts as a quality gate. The rewriter ensures retrieval precision. The router ensures the right source is queried. The validator ensures the output is grounded.
Individual failures get caught and corrected rather than silently propagated.
That said, the diagram below shows one of many blueprints of an Agentic RAG system. Production systems increasingly combine Corrective RAG, Adaptive RAG, Self-RAG, and hybrid search (vector + lexical with reranking) based on latency budgets and accuracy requirements.
👉 Over to you: What does your Agentic RAG setup look like?
@milan_milanovic It’s pleasure to connect. I am software engineer with 8+ years of experiences. Currently building https://t.co/g1HeRehEcj for tournament event!