--- title: "Muhammad Azlaan Zubair (@mdazlaanzubair) | Hashnode" description: "Software Engineer | Applied AI Researcher | AI Workflows & Automation. Read the latest articles by Muhammad Azlaan Zubair on Hashnode." url: "https://hashnode.com/@mdazlaanzubair" --- @mdazlaanzubair # Muhammad Azlaan Zubair @mdazlaanzubair · Karachi, Pakistan · Joined December 2022 Software Engineer | Applied AI Researcher | AI Workflows & Automation [Website](https://mdazlaanzubair.com/) ## Comments [Agreed. Abstention has to be learned from domain-specific ambiguity during RLCD training rather than magically inherited just by adding an "unknown" option. Thanks for pushing the epistemic boundary of the evaluation debate further!](https://hashnode.com/posts/jev-is-not-a-paradigm-shift-it-s-ust-a-fast-classifier/6ab6070e6f71e37fe1b72e1e/comment/6ab7974941d630990725fca0#reply-6ab7aa58f535934875946e7e) Reply · Article · Sep 26 · [Jev Is Not a Paradigm Shift. It’s ust a Fast Classifier.](https://hashnode.com/posts/jev-is-not-a-paradigm-shift-it-s-ust-a-fast-classifier/6ab6070e6f71e37fe1b72e1e) [Exactly. Retrieval quality and iteration speed are often more important than raw parameter count.](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7/comment/6ab04878b4d6167baf5a8b48#reply-6ab12c2eba08dafeab2bc6a0) Reply · Article · Sep 21 · [Does Size Matter? Bigger AI Isn’t Always Better](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7) [That’s right! AI is becoming increasingly accessible, and many models are already capable of handling everyday tasks. The real magic often lies in the systems around them, the prompts, context, tools, and workflows that shape their performance. That’s why the same Claude model can behave differently in Anthropic’s official app, Cursor, Antigravity, or through the API. The model matters, but how it’s integrated matters just as much.](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7/comment/6ab0fed4118f7fcc2354fa88#reply-6ab12be0ac7d9dc35fa3a2d1) Reply · Article · Sep 21 · [Does Size Matter? Bigger AI Isn’t Always Better](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7) [Quantizing a 405B model to 4-bit precision while maintaining full precision for smaller models is an unfair comparison due to outlier features and model size.](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7/comment/6aaf38148576fd0de007aa25#reply-6ab0252d072ecca869081910) Reply · Article · Sep 20 · [Does Size Matter? Bigger AI Isn’t Always Better](https://hashnode.com/posts/does-size-matter-bigger-ai-isn-t-always-better/6aaf0e4f4a3380a14d6989c7) [100% agree, Julian. The model provides the reasoning, but the loop/harness defines whether it actually succeeds or breaks in production. I recently wrote about this while analyzing leaked system prompts. Specifically looking at how production agents treat tool use as a hard operating contract (orchestration logic, failure handling, parallel execution) rather than simple prompt decoration: Are these really accidental leaks? Without that explicit loop logic around the model, even the best model degrades quickly once edge cases hit.](https://hashnode.com/posts/3-benchmark-the-model-harness-pair/6a8bf9b917de0e315d9dcde4/comment/6a8e85dbce5757b692289eba#reply-6a8e95bbc69942c3583976ee) Reply · Article · Aug 26 · [Benchmark the Model–Harness Pair, Not the Model](https://hashnode.com/posts/3-benchmark-the-model-harness-pair/6a8bf9b917de0e315d9dcde4) [Great callout. Ledger differences definitely add noise, and isolating accounting overhead from context growth is the right move for a cleaner analysis. Appreciate the feedback!](https://hashnode.com/posts/3-benchmark-the-model-harness-pair/6a8bf9b917de0e315d9dcde4/comment/6a8d16e88a9885c4cc1823ff#reply-6a8d64415cc32c7687b462b1) Reply · Article · Aug 25 · [Benchmark the Model–Harness Pair, Not the Model](https://hashnode.com/posts/3-benchmark-the-model-harness-pair/6a8bf9b917de0e315d9dcde4) [Exactly! In today’s competitive landscape, just having smart intelligence isn’t enough. What really matters is how well you build an ecosystem and how gracefully you handle things when limit exceeds.](https://hashnode.com/posts/claude-pro-worth-20/6a7790364ad9c2f665b30f76/comment/6a79d50a4b49af15de7f8943#reply-6a7a0ff1b17a9ebc546bcbb8) Reply · Article · Aug 10 · [Claude’s Intelligence Isn’t Worth $20 to Me Anymore](https://hashnode.com/posts/claude-pro-worth-20/6a7790364ad9c2f665b30f76) [So basically, it boils down to a trade-off between cost and security. Fair enough. That said, I feel like teams often default to massive models without evaluating whether a 3B or 7B parameter model can easily handle the workload. Since open-source models are built with general purpose scopes to chase benchmarks against top AI labs, businesses can dramatically right-size them by fine-tuning on their specific domain.](https://hashnode.com/posts/building-a-local-first-ocr-llm-pipeline-for-structured-business-documents/6a59006b1cb1447d23d1b10c/comment/6a7586158dd521a7a534b093#reply-6a799715f6150130b0b8e87e) Reply · Article · Aug 10 · 1 · [Building a Local-First OCR + LLM Pipeline for Structured Business Documents](https://hashnode.com/posts/building-a-local-first-ocr-llm-pipeline-for-structured-business-documents/6a59006b1cb1447d23d1b10c) [Great read on the local-first architecture! One thing I'm wondering about though: doesn't shifting to locally hosted OCR VLMs and serving models locally significantly increase your infrastructure and inference costs compared to lightweight APIs, especially if you have to provision dedicated GPUs?](https://hashnode.com/posts/building-a-local-first-ocr-llm-pipeline-for-structured-business-documents/6a59006b1cb1447d23d1b10c/comment/6a7586158dd521a7a534b093) Comment · Article · Aug 7 · 1 · [Building a Local-First OCR + LLM Pipeline for Structured Business Documents](https://hashnode.com/posts/building-a-local-first-ocr-llm-pipeline-for-structured-business-documents/6a59006b1cb1447d23d1b10c) [Exactly, Mateo Ruiz. The caching parallel is spot on; it's less about raw storage capacity and much more about efficient eviction and prioritization policies as the horizon stretches. Thanks for reading!](https://hashnode.com/posts/ai-agents-context-budget/6a71f319030746384a2fcb62/comment/6a7325359a8f308a41c76cfb#reply-6a73a8b2ed41826978a89383) Reply · Article · Aug 5 · [AI Agents Do Not Just Need More Context. They Need a Context Budget](https://hashnode.com/posts/ai-agents-context-budget/6a71f319030746384a2fcb62)