Kaushal Prajapati writes about production AI engineering, agentic systems, software engineering, deep-learning papers, evals, retrieval, and LLM platforms.https://kforcode.dev/en-ushttps://kforcode.dev/blog/thousand-tools-one-backpack/https://kforcode.dev/blog/thousand-tools-one-backpack/Put Codex, Claude Code, Hermes Agent, and DeepSeek-style Code Mode beside the same hypothetical 50,000-tool company and compare what actually reaches the model.Wed, 09 Sep 2026 00:00:00 GMTagentsharnessmcpskillscontext-engineeringhttps://kforcode.dev/blog/open-model-accuracy/https://kforcode.dev/blog/open-model-accuracy/Open model accuracy is also a systems problem. Hermes Agent narrows tool choices, repairs safe mistakes, preserves context, and helps smaller models finish.Wed, 02 Sep 2026 00:00:00 GMTagentsharnessevalsopen-sourcehttps://kforcode.dev/blog/hermes-agent-architecture/https://kforcode.dev/blog/hermes-agent-architecture/The hermes-agent architecture at a glance: why the largest open-source agent harness ships both of its cheap context reducers switched off, and what that buys.Mon, 31 Aug 2026 00:00:00 GMTagentsharnesscontextcosthttps://kforcode.dev/blog/hermes-agent-internals/https://kforcode.dev/blog/hermes-agent-internals/A component-by-component study of hermes-agent internals: the loop, the prompt, the cache plan, compaction, state, approvals, and what each one costs.Mon, 31 Aug 2026 00:00:00 GMTagentsharnesscontextcosthttps://kforcode.dev/blog/the-system-layer-behind-ai-products/https://kforcode.dev/blog/the-system-layer-behind-ai-products/AI demos are easy; AI products are hard. The gap is a runtime — the shared execution layer that handles model routing, tool authorization, long-context state, memory, and fallback. Here is what it takes to run one at scale.Mon, 20 Jul 2026 00:00:00 GMTagentsai-platformruntimellmproductionhttps://kforcode.dev/blog/evidence-retrieval-86-to-97/https://kforcode.dev/blog/evidence-retrieval-86-to-97/Compliance answers are only trustworthy if every claim cites the right evidence. Here is how hybrid retrieval, reranking, control-aware chunking, and citation-span ranking took evidence F1@10 from 86% to 97%.Tue, 14 Jul 2026 00:00:00 GMTragretrievalevalscompliancellmhttps://kforcode.dev/blog/one-model-instead-of-23/https://kforcode.dev/blog/one-model-instead-of-23/A model per document type does not scale. Here is how a single layout-aware extraction architecture replaced 23+ document-specific models, cut OCR spend 70%, and made new document types cheap to add.Tue, 07 Jul 2026 00:00:00 GMTdocument-aiocrml-systemscostarchitecture