Ankush Patel
https://ankushp.com/blog
ML engineer pioneering human-centered AI solutionsen-us[email protected] (Ankush Patel)[email protected] (Ankush Patel)Mon, 09 Feb 2026 00:00:00 GMThttps://ankushp.com/blog/posts/memory-state-and-skills-are-not-the-same-thingMemory, State, and Skills Are Not the Same Thing
https://ankushp.com/blog/posts/memory-state-and-skills-are-not-the-same-thing
Memory, state, and skills solve different problems in AI systems. Blurring them leads to muddled architecture, confusing product behavior, and weaker workflows.Mon, 09 Feb 2026 00:00:00 GMT[email protected] (Ankush Patel)memorysystem-designtoolinghttps://ankushp.com/blog/posts/assistant-workflow-or-agent-the-boundary-actually-mattersAssistant, Workflow, or Agent? The Boundary Actually Matters
https://ankushp.com/blog/posts/assistant-workflow-or-agent-the-boundary-actually-matters
A practical framing for separating assistants, workflows, and agents based on control, execution latitude, and architecture.Thu, 22 Jan 2026 00:00:00 GMT[email protected] (Ankush Patel)product-boundariesagentic-workflowsorchestrationhttps://ankushp.com/blog/posts/designing-personal-ai-automations-that-people-actually-trustDesigning Personal AI Automations That People Actually Trust
https://ankushp.com/blog/posts/designing-personal-ai-automations-that-people-actually-trust
Personal AI automations become trustworthy when scope, timing, approvals, and failure behavior are easy to understand and control.Fri, 05 Dec 2025 00:00:00 GMT[email protected] (Ankush Patel)automationreliabilityproduct-boundarieshttps://ankushp.com/blog/posts/how-i-think-about-reliability-in-agentic-workflowsHow I Think About Reliability in Agentic Workflows
https://ankushp.com/blog/posts/how-i-think-about-reliability-in-agentic-workflows
A short post on reliability in agentic workflows: make mistakes visible, bounded, and recoverable instead of optimizing for intelligence theater.Wed, 19 Nov 2025 00:00:00 GMT[email protected] (Ankush Patel)reliabilityagentic-workflowssystem-designhttps://ankushp.com/blog/posts/memory-is-useful-but-only-after-the-workflow-is-stableMemory Is Useful — But Only After the Workflow Is Stable
https://ankushp.com/blog/posts/memory-is-useful-but-only-after-the-workflow-is-stable
Memory helps only after a workflow is clear and reliable. Otherwise it preserves inconsistency and makes a shaky AI system harder to understand.Tue, 07 Oct 2025 00:00:00 GMT[email protected] (Ankush Patel)memoryagentic-workflowssystem-designhttps://ankushp.com/blog/posts/the-missing-layer-in-your-ai-stack-intermediate-knowledgeThe Missing Layer in Your AI Stack: Intermediate Knowledge
https://ankushp.com/blog/posts/the-missing-layer-in-your-ai-stack-intermediate-knowledge
Most AI products hit a wall at scale. The solution is not just caching responses—it is building an Intermediate Knowledge Layer that stores the ingredients feeding your prompts.Mon, 29 Sep 2025 00:00:00 GMT[email protected] (Ankush Patel)aiarchitecturecachingllmhttps://ankushp.com/blog/posts/tool-calling-is-the-real-product-surface-of-ai-systemsTool Calling Is the Real Product Surface of AI Systems
https://ankushp.com/blog/posts/tool-calling-is-the-real-product-surface-of-ai-systems
AI products become truly useful through their tool layer. The real product surface is not the chat UI, but the actions the system can safely and reliably take.Thu, 21 Aug 2025 00:00:00 GMT[email protected] (Ankush Patel)toolingsystem-designorchestrationhttps://ankushp.com/blog/posts/why-most-ai-workflows-dont-need-multiple-agentsWhy Most AI Workflows Don’t Need Multiple Agents
https://ankushp.com/blog/posts/why-most-ai-workflows-dont-need-multiple-agents
Most AI workflows should start as clear, controlled workflows with tools and explicit state. Multi-agent structure only makes sense when specialization or parallelism truly helps.Wed, 11 Jun 2025 00:00:00 GMT[email protected] (Ankush Patel)agentic-workflowsorchestrationsystem-designhttps://ankushp.com/blog/posts/the-mlops-engineer-bridging-the-gap-between-machine-learning-and-productThe MLOps Engineer: Bridging the Gap Between Machine Learning and Product
https://ankushp.com/blog/posts/the-mlops-engineer-bridging-the-gap-between-machine-learning-and-product
MLOps is essential for taking machine learning models from development to production. This article covers its role in the AI ecosystem, key challenges it addresses, and the tools and practices that make it possible to deploy, scale, and monitor ML systems effectively.Wed, 04 Dec 2024 00:00:00 GMT[email protected] (Ankush Patel)mlopsmachine-learningtoolshttps://ankushp.com/blog/posts/accelerating-your-python-projects-with-poetryAccelerating Your Python Projects with Poetry
https://ankushp.com/blog/posts/accelerating-your-python-projects-with-poetry
10x your Python project developer experience with Poetry - example project and Poetry workflow included!Mon, 18 Mar 2024 00:00:00 GMT[email protected] (Ankush Patel)pythontools