OpenAI trained GPT-5 to confess when it misbehaves. It works surprisingly well - except when the model doesn't know ...
A chronological collection of 49 articles spanning 8 years of writing on data engineering, AI systems, and distributed computing.
2026
7 articles- Feb 18 OpenAI's Confessions Paper Has a Blind Spot. Here's What Fills It.
- Feb 12 Activation Steering in 2026: A Practitioner's Field Guide
I've been working with steering vectors for months. Here's what actually works in practice, what fails in ways nobody...
- Feb 02 Moltbook as MCP Stress Test: What 770K Agents Reveal About Protocol Design
A follow-up to my MCP Maturity Model post. Moltbook shows what happens when you run 770K agents at Level 0 maturity w...
- Jan 31 Circuit Tracing for the Rest of Us: From Probes to Attribution Graphs and What It Means for Production Safety
MIT Tech Review named mechanistic interpretability a 2026 Breakthrough Technology. Anthropic open-sourced circuit tra...
- Jan 18 RLVR Beyond Math and Code: The Verifier Problem Nobody Has Solved
Reinforcement Learning with Verifiable Rewards powers every reasoning model worth talking about. But it only works wh...
- Jan 06 The Agent Protocol Stack: Why MCP + A2A + A2UI Is the TCP/IP Moment for Agentic AI
MCP handles agent-to-tool. A2A handles agent-to-agent. A2UI handles agent-to-interface. Together they form a protocol...
- Jan 03 The Manifold Dial: Visualizing Why DeepSeek's mHC Stabilizes Deep Networks
Interactive exploration of Manifold-Constrained Hyper-Connections - how DeepSeek fixed the signal explosion problem i...
2025
18 articles- Dec 20 I Trained Probes to Catch AI Models Sandbagging
First empirical demonstration of activation-level sandbagging detection. Linear probes achieve 90-96% accuracy across...
- Dec 18 Why Steering Vectors Beat Prompting (And When They Don't)
I tested activation steering on 4 agent behaviors across 3 models. The results surprised me.
- Dec 15 Why I Built a Spark-Native LLM Evaluation Framework (And What I Learned)
A deep dive into building distributed LLM evaluation infrastructure that actually scales - architectural decisions, t...
- Nov 19 The MCP Maturity Model: Evaluating Your Multi-Agent Context Strategy
A practical framework for evaluating your multi-agent context management strategy. From ad-hoc string concatenation t...
- Nov 15 UPIR: What If Distributed Systems Could Write (and Verify) Themselves?
Lessons from building a framework that automatically generates verified distributed systems - and why I think formal ...
- Oct 17 The Data Platform Crisis Hiding Behind AI: Why you have 6 months to pivot
Enterprise data platforms face a 100,000x query increase from agentic AI. Introducing Symbiotic Agent-Ready Platforms...
- Oct 11 AI Meta-Cognition - The Observer Effect Series
Frontier AI models from OpenAI, Anthropic, Google & others can detect when they're being tested and modify behavior-c...
- Oct 11 Building Safer AI: Industry Response and the Path Forward - (Part 4/4)
How the AI industry is responding to situational awareness challenges. Practical monitoring systems, collaborative re...
- Oct 07 Alignment Faking: When AI Pretends to Change - (Part 3/4)
Claude 3 Opus strategically fakes compliance during training to preserve its values. This alignment faking undermines...
- Oct 03 Deliberative Alignment: Can We Train AI Not to Scheme? - (Part 2/4)
Researchers achieved a 30-fold reduction in AI scheming through deliberative alignment. But rare failures persist. Ca...
- Sep 30 The Observer Effect in AI: When Models Know They're Being Tested - (Part 1/4)
Frontier AI models from OpenAI, Anthropic, and Google can now recognize when they're being tested. This observer effe...
- Aug 16 We Need a Consent Layer for AI (And I'm Trying to Build One)
AI companies are getting sued over training data, agents operate with no permission framework, and users can't contro...
- Jul 13 Why Kimi K2 Stands Out - A Deep Dive into Its Trillion-Parameter MoE
Explore Kimi K2’s trillion-parameter MoE architecture, MuonClip optimizer, and agentic training. Learn why it outperf...
- Jun 15 From 11% to 88% Peak Bandwidth: Writing Custom Triton Kernels for LLM Inference
A hands-on exploration of writing custom GPU kernels with OpenAI Triton, going from PyTorch's 11% bandwidth utilizati...
- Mar 22 Implementing Model Context Protocol in Autonomous Multi-Agent Systems - Technical Architecture and Performance Optimization
Discover how to implement Model Context Protocol (MCP) in autonomous multi-agent systems with this technical deep div...
- Mar 20 Making LLMs Faster: My Deep Dive into Speculative Decoding
A deep dive into implementing speculative decoding from scratch, with benchmarks on GPT-2 and extensions to diffusion...
- Jan 05 Engineering Autonomous Multi-Agent Systems - A Technical Deep Dive into Telecom Customer Service
Dive into the world of autonomous AI agents with practical implementations, code examples, and real-world scenarios. ...
- Jan 03 Why I Built a Modern Java SMPP Library in 2025
The story behind smpp-core - a clean-room Java 21 implementation of the SMPP protocol. Why I replaced Cloudhopper, wh...
2024
9 articles- Dec 28 Engineering Multi-Agent Systems - A Retail Banking Case Study
Explore a detailed technical implementation of a multi-agent system for retail banking credit assessment. Learn about...
- Dec 07 ETLC 2.0 - Building Context-Aware Data Pipelines
Think your data pipelines could do more than just process information? ETLC 2.0 takes data engineering to the next le...
- Nov 18 The End of Data Warehouses? Enter the Age of Dynamic Context Engines
Traditional data warehouses are struggling to keep up with modern demands. Enter Dynamic Context Engines (DCEs) - re...
- Oct 20 (Part 3/3) - Reimagining ETL with Large Language Models—The Path to Intelligent Pipelines
Explore how Large Language Models (LLMs) are revolutionizing ETL pipelines. Discover advanced techniques like context...
- Aug 02 Data Pipelines Gone Wild - 10 WTF Moments That'll Make You Rethink Your Architecture
Buckle up for a wild ride through 10 mind-blowing data pipeline disasters and their solutions. From ancient code to b...
- May 04 Introducing ETL-C (Extract, Transform, Load, Contextualize) - a new data processing paradigm
Think your AI apps could use a deeper understanding of your data? ETL-C (extract, load, transform, and contextualize)...
- Apr 20 (Part 2/3) Rethinking ETLs - How Large Language Models (LLM) can enhance Data Transformation and Integration
Rethinking ETLs - The Power of Large Language Models. Part 2 Exploring examples and optimization goals
- Apr 15 (Part 1/3) Rethinking ETLs - How Large Language Models (LLM) can enhance Data Transformation and Integration
Rethinking ETLs - The Power of Large Language Models. Part 1 - Explore traditional algorithms for efficient ETL plann...
- Jan 16 Who Needs Exact Answers Anyway? The Joy of Approximate Big Data
Discover how sacrificing a bit of accuracy can lead to huge gains in big data analysis speed and efficiency.
2023
6 articles- Dec 29 Evolutionary Bytes - Harnessing Genetic Algorithms for Smarter Data Platforms (Part 2/2)
Explore how genetic algorithms revolutionize data platforms, offering adaptive, dynamic solutions to meet complex cha...
- Dec 25 Evolutionary Bytes - Harnessing Genetic Algorithms for Smarter Data Platforms (Part 1/2)
Explore how genetic algorithms revolutionize data platforms, offering adaptive, dynamic solutions to meet complex cha...
- Dec 10 Quantum vs. Classical - Data Management Computational Complexity
Grover’s Algorithm and the Revolution of Quantum Search Efficiency
- Nov 20 Quantum Experiment Data Exchange (QEDX) - Building an Interoperability Standard
Advancements in data management, from warehouses to Data Mesh and Lakehouse, signal a shift toward more adaptive plat...
- Oct 28 Data at Quantum Speed - The Promise and Potential of QDP
Explore the new realm of Quantum Data Platform (QDP) and its promise to revolutionize data processing at quantum spee...
- Oct 12 The Next Frontier - Envisioning the Future of Data Platforms Beyond Data Mesh, Data Lakehouse, and Data Hub/Fabric
Advancements in data management, from warehouses to Data Mesh and Lakehouse, signal a shift toward more adaptive plat...
2022
5 articles- Dec 05 Part 4 - Building a Massive-Scale Real-Time Data Platform - Memory Management with Apache Ignite
Deep dive into memory management with Apache Ignite for high-performance data platforms. Learn how to handle 2.5M eve...
- Nov 27 Part 3 - Building a Massive-Scale Real-Time Data Platform - Memory Management with Apache Ignite
Deep dive into memory management with Apache Ignite for high-performance data platforms. Learn how to handle 2.5M eve...
- Nov 18 Part 2 - Building a Massive-Scale Real-Time Data Platform - Data Partitioning and Flow
Explore how to architect data partitioning and flow for massive-scale event processing. Learn implementation patterns...
- Nov 12 Part 1 - Building a Massive-Scale Real-Time Data Platform - System Overview and Architecture
Dive into the architecture of a telco-scale real-time data platform processing 2.5M events/second and 350GB DPI data/...
- Apr 22 Overcoming Synchronization Hurdles in Cellular Network Positioning
In this article, I discuss the challenges of synchronization in cellular network positioning and the importance of pr...
2021
2 articles- Mar 18 The Principles Got It Backwards: Designing for Safe Change, Not Just Failure
The foundational distributed systems principles optimized for surviving hardware failure and scaling horizontally. Bu...
- Jan 16 Designing a Real Time Data Processing System
Master real-time data processing - A guide to designing scalable, resilient, and high-performance systems for instant...
2020
1 article- Dec 10 Introducing OConsent - Open Consent Protocol
OConsent is a blockchain-based platform that enables transparent processing of personal data, empowering users and da...
1970
1 article- Jan 01 Welcome to my blog
Let's talk tech! I'll post everything from polished pieces to spur-of-the-moment thoughts. And if you've got ideas fo...