Self-updating service

A curriculum that never goes stale.

Once per week, a frontier AI agent, running with maximum thinking, researches what has changed, trended or become obsolete in AI engineering, then evolves this curriculum: adding, updating and retiring tracks, concepts, projects and resources under automated safeguards. This page reports the most recent run, full transparency, real data.

Latest run

Most recent self-update.

Last result: updated
Last update date (IST): 2026-W36
Last run at: 2026-08-30T19:01:43.774Z
Curriculum version: 12
Note: Updated agentic, modeling, and production evaluation concepts to incorporate Reinforcement Learning from Verifiable Rewards (RLVR), Model Context Protocol (MCP) standards, and Evaluation-Driven Development (EDD). (4 change(s) applied). via gemini-3.7-flash (max thinking).

Changelog

Recent curriculum changes.

Every weekly run is logged. The curriculum evolves with the frontier, and you can see exactly when.

Most recent changelog entries, newest first
Date	Version	Summary
2026-08-30	v12	Updated agentic, modeling, and production evaluation concepts to incorporate Reinforcement Learning from Verifiable Rewards (RLVR), Model Context Protocol (MCP) standards, and Evaluation-Driven Development (EDD). (4 change(s) applied).
2026-08-23	v11	Update curriculum concepts across Agentic Systems, AI Infrastructure, and Applied ML to incorporate Agent-to-Agent (A2A) protocol standardization, Process Reward Models (PRMs), persistent radix prefix caching, and synthetic mid-training capability injection. (4 change(s) applied).
2026-08-16	v10	Integrate GRPO/RLVR reasoning post-training, Multi-Head Latent Attention (MLA), Prefill-Decode (PD) Disaggregation, Test-Time Compute scaling, and Hybrid RAG into the curriculum. (5 change(s) applied).
2026-07-26	v9	Update MCP and tool-use concepts in the Agentic Systems track to align with the stateless Model Context Protocol specification and pre-compiled toolmaking architectures. (2 change(s) applied).
2026-07-20	v8	Evolve the curriculum to focus on Harness Engineering, JIT Context Hubs, outer-loop verification, and async cloud-native sandboxes, incorporating key June-July 2026 research findings and model economics. (8 change(s) applied).
2026-07-12	v7	Evolved the curriculum to integrate modern mechanistic auditing (J-space/J-lens), formal Lean 4 verification, real-to-sim world modeling, skill gates, counterfactual evaluations, and the latest sovereign AI regulatory sandboxes. (2 change(s) applied, 6 skipped by guardrails).
2026-07-06	v6	Evolve the curriculum to support Model Context Protocol (MCP) stateful interoperability, Software 3.0 compilation (SkillOpt/DSPy), stateful API architectures with managed sandboxes, and modern hybrid SSM-Transformer architectures. (7 change(s) applied).
2026-06-28	v5	Evolve the curriculum to incorporate ASIC LLM co-design, deterministic agent trajectory evaluation, dynamic multi-provider routing for resiliency, and alternative architectures (SSMs/Mamba) to bypass single-provider vulnerabilities and coordinate agents under formal protocols. (1 change(s) applied, 6 skipped by guardrails).