Weekly Tech News and Project Notes
This site does two things. It publishes weekly news digests — short snippets on AI and LLMs, on data and orchestration, and on breakthroughs in mathematics, CS and physics — researched and posted on a schedule by an automated pipeline, with me reviewing before anything goes live. It also collects links to my blog posts about the projects in my GitHub repositories: GraphRAG, agentic design patterns, and the ENGRAM memory system.
The reference pages further down began as a personal knowledge base. ENGRAM has taken over that job, but the pages are still worth keeping around as concise guides to the frameworks and databases I work with, and I add to them occasionally.
News
Weekly snippets, collected automatically:
My Blog Posts on Igor’s Musings About Technology
Blog posts about projects in my GitHub repository:
- Something Smaller Than an Agent: ENGRAM Scopes Authority to a Shared Project
- When the Question Has a Right Answer: ENGRAM Allows Agents to Work from a Brief
- What the Transcript Couldn’t Tell Me: ENGRAM Learns to Read What Actually Shipped
- No Message Passed Between Them: ENGRAM Agents Coordinate Through Memory
- What I Believed Last Tuesday: ENGRAM Learns to Change Its Mind
- The Layer a Harness Can’t Own
- A Team of Its Own: ENGRAM Agents Learn to Grow and Orchestrate a Team
- Standing Up Your Own Agents: ENGRAM Adds an Agent Manager Role
- A Brain Between Sessions — and Between Agents: ENGRAM Learns to Recall on Demand
- Memories Worth Sharing: ENGRAM Learns to Be Told What to Keep – and What to Share
- Where Memories Come From: ENGRAM Learns to Track Provenance and Group by Project
- A Mind That Knows What It’s Missing: ENGRAM Now Finds and Closes Its Own Gaps
- Completing the Mind: ENGRAM Now Remembers Conversations from Claude Desktop, ChatGPT, and Gemini
- Claude Code and ENGRAM Knowledge Hub: recalling the good memories together
- ENGRAM Knowledge Hub: A Personal Knowledge Graph That Grows With Your Research
- When Graphs Remember Better Than Summaries
- ENGRAM Part 3: Hippocampal-Inspired Memory – Working Memory, Long-Term Memory, and Periodic Consolidation
- ENGRAM Part 2: Unified Graph Retrieval from Long Documents and Multi-Agent Response Synthesis
- ENGRAM Part 1: GraphRAG with Hippocampal-Like Associative Retrieval
- MCP, A2A, and Human-in-the-Loop: A Multi-Agent Threat Intelligence System in Practice
- A Framework for Testing Agentic Design Patterns Using LangChain & LangGraph
- AI Agent with RAG-Assisted Declarative Query Generation for Data in Oracle Analytic Views
- Building a GraphRAG System – Core Infrastructure & Document Ingestion
- GraphRAG Part 2 – Cross-Doc & Sub-graph Extraction, Multi-Vector Entity Representation
- GraphRAG Part 3 – Intelligent MVR, Query Routing and Context Generation
- GraphRAG Part 4 – Community Detection and Embedding, Search and Hybrid Retrieval Integration
- Beyond Vector Search: How GraphRAG Enables Smarter AI Responses
- Workflow and Agentic Flow Patterns: Choosing the Right Approach
- The Shifting Landscape of Software Development Roles in the AI Era
- The Hybrid Approach: Bridging Traditional and Agentic Workflows
- RAG at Scale: The Hidden Trade-Offs of Single Vector Embedding
- From Static Rules to Smart Agents: Advent of Agentic Workflows
- Why LLMs and Humans Think Differently: An Amateur Dive into AI’s Compression Problem
Knowledge Base Pages
General Topics:
Google AI & Agent Developmnet Kit (ADK):
- Google ADK (Agent Development Kit) — Concise Guide
- Inter-application Agents Communication with Google A2A
- Google Vertex AI - Concise Guide
LangChain / LangGraph Framework:
- LangChain & LangGraph — Concise Guide
- Intra-Application Agent Communication in LangGraph
- LangChain Core: Prompts and Runnables for Agent Flows
Neo4J Graph Database:
Oracle Graph Database:
Various:
- 2026 Major Breakthroughs in Mathematics, so far
- 2025 Major Brteakthroughs in Mathematics
- Humanity and Powerful AI
- Various Tips and How To Dos…