I build agentic solutions, from small companies to enterprises.https://juliardifh.com/enhttps://juliardifh.com/writing/thoughts/2026-09-19-orchestration-choreography-agents/https://juliardifh.com/writing/thoughts/2026-09-19-orchestration-choreography-agents/Software engineers often look at AI and assume their experience is about to be replaced. It is not. Microservices and AI agent systems use the same two architecture patterns, orchestration and choreography. What changed is the actor: a service is deterministic, an agent is probabilistic, and every difference between the two worlds comes from that.Sat, 19 Sep 2026 00:00:00 GMTagentsarchitectureplatformhttps://juliardifh.com/writing/thoughts/2026-09-harness-overkill/https://juliardifh.com/writing/thoughts/2026-09-harness-overkill/Harnesses like OpenClaw, Hermes Agent and Deep Agents are built for a general assistant. A harness has four parts: tools, memory, control flow and evaluation. A specialised agent still needs all four, but a narrow slice of each.Thu, 17 Sep 2026 00:00:00 GMTagentsharnessllmhttps://juliardifh.com/writing/thoughts/2026-06-on-prem-makin-masuk-akal/https://juliardifh.com/writing/thoughts/2026-06-on-prem-makin-masuk-akal/Lanjutan dari hitungan balik modal sebelumnya. Rupiah yang melemah memajukan balik modal dari bulan ke-7 ke bulan ke-5, model kecil seperti DharmaOCR mengalahkan API frontier di bidang yang sempit, dan model lokal diperbarui hampir tiap bulan.Thu, 18 Jun 2026 12:00:00 GMTllmon-premroihttps://juliardifh.com/writing/thoughts/2026-06-roi-llm-on-premise/https://juliardifh.com/writing/thoughts/2026-06-roi-llm-on-premise/Dua alasan LLM on premise makin masuk akal: data sensitif tidak pernah keluar dari mesin sendiri, dan biaya tidak ikut meledak saat rupiah melemah. Lengkap dengan rumus y = wx + b, contoh hitungan di tiga volume percakapan, dan daftar biaya yang belum masuk hitungan.Thu, 18 Jun 2026 10:00:00 GMTllmon-premroihttps://juliardifh.com/writing/thoughts/2026-03-context-engineering/https://juliardifh.com/writing/thoughts/2026-03-context-engineering/Three things kill agent projects, and none of them is the framework: cost, hallucination and the context window. All three come from what goes into the prompt. Five techniques that cut the bill on one project by about 40% by sending the right context instead of more of it.Wed, 18 Mar 2026 00:00:00 GMTllmagentscontexthttps://juliardifh.com/writing/thoughts/2026-02-rethinking-langgraph/https://juliardifh.com/writing/thoughts/2026-02-rethinking-langgraph/LangGraph solved a real coordination problem in 2024. Now that tool calling and agent-to-agent messaging have open protocols, the graph starts to feel like a flowchart wrapped around an intelligent model.Wed, 18 Feb 2026 14:00:00 GMTagentslanggraphmcphttps://juliardifh.com/writing/thoughts/2026-02-how-agents-talk/https://juliardifh.com/writing/thoughts/2026-02-how-agents-talk/Agent communication was standardised in the late nineties and failed commercially for being too formal. The three camps today, routers, LLM routing and protocols like A2A and MCP, are rediscovering what FIPA got right.Wed, 18 Feb 2026 13:00:00 GMTagentsmcpprotocolshttps://juliardifh.com/writing/thoughts/2026-02-renting-a-gpu/https://juliardifh.com/writing/thoughts/2026-02-renting-a-gpu/Nvidia's speech-to-speech model would not run on an M4 MacBook Pro, so I rented an A40 from RunPod. Managing rental GPUs, cost per minute and knowing when to tear them down, is a skill of its own.Wed, 18 Feb 2026 12:00:00 GMTmlopsgpuvoicehttps://juliardifh.com/writing/thoughts/2026-02-openclaw-as-jarvis/https://juliardifh.com/writing/thoughts/2026-02-openclaw-as-jarvis/Wired into Telegram on a sandboxed VM, it answered a finance-team question across business, operations and technical dimensions at once. Also why this kind of tool stays in a virtual machine.Wed, 18 Feb 2026 11:00:00 GMTagentsllmsecurityhttps://juliardifh.com/writing/thoughts/2026-02-dry-go-protocol-buffers/https://juliardifh.com/writing/thoughts/2026-02-dry-go-protocol-buffers/The protocol buffer already describes the domain, so redeclaring the same fields in Go duplicates the model. Referencing and embedding the generated types instead, and what that costs.Fri, 06 Feb 2026 00:00:00 GMTgogrpcprotobufhttps://juliardifh.com/writing/thoughts/2025-12-ai-in-the-workplace/https://juliardifh.com/writing/thoughts/2025-12-ai-in-the-workplace/The framework from a talk on AI in the workplace: start from expensive business problems rather than trends, match cloud, on-premise or hybrid to your constraints, and implement when the business impact is clear.Thu, 18 Dec 2025 00:00:00 GMTllmtalkslifehttps://juliardifh.com/writing/thoughts/2024-10-vector-database-opensearch-tfidf/https://juliardifh.com/writing/thoughts/2024-10-vector-database-opensearch-tfidf/Implementing a vector database on OpenSearch with TF-IDF in Python: what a vector store is for, how the index is built, and how the search behaves.Fri, 18 Oct 2024 00:00:00 GMTaipythonsearchhttps://juliardifh.com/writing/thoughts/2019-10-goroutines-broadcast-campaign/https://juliardifh.com/writing/thoughts/2019-10-goroutines-broadcast-campaign/How a marketing broadcast at Sorabel went nine times faster once the work was split across goroutines with bigger payloads, and what that taught me about batching.Tue, 15 Oct 2019 00:00:00 GMTgo