Leksh·Aug 21How Cleary Is Built: A Real MCP Application (Part 2/3)In Part 1, we looked at the accessibility problems Cleary solves. In this post, we’ll look under the hood at how it’s built.
Leksh·Jul 21Deploying on Alibaba Cloud’s Free Tier (Part 4 of 4)Part of a four-part series on designing and evaluating a research assistant with structured memory.
Leksh·Jul 21MemoryBench: Proving It Works (Part 3 of 4)Part of a four-part series on designing and evaluating a research assistant with structured memory.
Leksh·Jul 20Building the Memory Engine (Part 2 of 4)Part of a four-part series on designing and evaluating a research assistant with structured memory.
Leksh·Jul 20Building Memory That Can Change Its Mind (Part 1 of 4)A four-part series on designing a research assistant with memory that behaves less like a transcript and more like a careful human…
Leksh·Jul 16[NLP Series] How Do Machines Know Two Words Are Similar? Understanding Cosine SimilarityIn the previous posts, we learned how machines represent words as vectors using embeddings. But once every word becomes a vector, another…
Leksh·Jun 30Gen AI Academy Day 6: Evaluating and Testing AI Agents with Google’s ADKBuilding an AI agent is only half the journey. Ensuring it behaves reliably is what makes it production-ready.
Leksh·Jun 16Gen AI Academy Day 5: Mastering Agent Workflows in Google’s ADKHow do AI agents decide who should do what, in what order, and when to stop?
Leksh·Jun 15Gen AI Academy Day 4: Designing Multi-Agent Systems with Google’s ADKWhy use one agent when a team of specialised agents can solve the problem better?
![[NLP Series] How Do Machines Know Two Words Are Similar? Understanding Cosine Similarity](https://webproxy.poorya-velaei-d67.workers.dev/https://miro.medium.com/v2/resize:fill:160:106/1*p44AiTsIvLDymohYxlZ9XQ.png)
![[NLP Series] How Do Machines Know Two Words Are Similar? Understanding Cosine Similarity](https://webproxy.poorya-velaei-d67.workers.dev/https://miro.medium.com/v2/resize:fill:320:214/1*p44AiTsIvLDymohYxlZ9XQ.png)
