Expert profile
Founder & AI Product Leader
Yash Thakker is a Generative AI expert with over 12 years of experience in product leadership and technical strategy. As the founder of explainx.ai, he has taught over 300,000 learners and built AI platforms serving millions of users globally. He specializes in Agentic AI, Multimodal RAG, and the intersection of LLMs with consumer hardware.
llms.txt is an open specification for providing LLM-friendly markdown content at /llms.txt. Learn how this simple standard helps AI assistants like ChatGPT,...
OpenAI MRC: multipath GPU networking (RoCE, packet spraying, SRv6) for frontier training; OCP spec. Diagrams from OpenAI’s post; LLM tokens vs fabric...
Traditional RAG relies on vector databases, embeddings, and chunking. Agentic RAG uses primitive search tools and structured traversal. Learn why Claude...
Recursive models trade additional internal computation for potential reasoning gains, which must be evaluated alongside latency and task generalization.
Astrocade’s May 2026 round totals $56M (Series B: Sequoia; Series A: Sea) with NVIDIA, Google AI Futures Fund, LG Tech Ventures & more—per the company blog.
Browserbase skills for Claude Code: remote browse, bb CLI, traces, cookie-sync, safe-browser—npx skills add browserbase/skills or browse@browserbase plugin.
CocoIndex (Apache-2): Rust core + Python API—incremental delta embeddings to Postgres for agent RAG. pip install cocoindex; github.com/cocoindex-io/cocoindex.
Deep tutorial for OpenAI Codex desktop pets: Settings, /pet overlay, hatch-pet install, sprite prompts, troubleshooting, security. Plus archetypes for best...
MCP context-mode: sandbox bulky tool output + SQLite session FTS for agents; Claude Code plugin or npx. Elastic License v2. github.com/mksglu/context-mode.
DeepSeek-TUI (MIT): Rust terminal agent for DeepSeek V4 Pro/Flash—MCP, skills, auto routing, HTTP serve. Third-party harness: github.com/Hmbown/DeepSeek-TUI.
Use Maigret to discover public username matches, then verify identity and affiliation separately before turning results into conclusions.
Subquadratic’s SubQ pairs a sub-quadratic sparse-attention stack (SSA) with a 12M-token positioning; official SSA post cites 52× prefill at 1M vs dense FA...
