Agent Drift: How Autonomous AI Agents Lose the Plot
Autonomous AI agents degrade over long-horizon tasks through accumulated context pollution, not single failures. What agent drift is, why it happens, and how to architect against it.
Read more →Blog
Technical insights on machine learning, MLOps, AI engineering, and software development. Sharing practical experiences and deep dives into the latest trends in tech.
Browse by Series
The AI Hourglass
The AI stack is shaped like an hourglass — wide at energy and applications, pinched at chips and GPUs. But the pinch is shifting. Everyone's watching Nvidia; the real chokepoint is one layer below.
AI Semiconductors Geopolitics
Read more →
How to Build AI Agents: Inside Out
Build AI agents from the inside out — start with behavior in CLAUDE.md, add capabilities via MCP, then wrap in code with the Claude Agent SDK.
Read more →
Repowire: Multi-Agent Coding Workflow Across Repos
Repowire is a pull-based mesh network that lets Claude Code sessions communicate across repositories in real-time, replacing stale docs with live agent-to-agent queries.
Read more →
The Vibe Bottleneck: AI Coding Workflow Across Multi-Repo
Multi-repo coding agent workflows turn you into the message bus. When work spans repositories, you become the slowdown and the source of errors.
Read more →
Agentic Cloud Infrastructure: Beyond 12-Factor Apps
Agents are stateful logic, not stateless apps. The cloud primitives — durable execution, agent identity, MCP — that AI agents need to run reliably.
Read more →
AI Agent Memory Management With Databases
AI agent memory management mirrors database design. Agent state needs persistence, retries, and checkpointing — the same problems backend systems solved.
Read more →
AI and the Future of Programming: Architects of Intent
Code is becoming a transient artifact — AI-generated bytecode. The future of programming belongs to Architects of Intent, not writers of loops.
Read more →
Ambient AI: Building Torale from Complexity to Simplicity
Torale monitors the web so you don't have to. How ambient AI turns passive monitoring into proactive intelligence — and how Gemini made it possible.
Read more →
ML Infrastructure Strategy: Build for Momentum
Early architectural decisions create a flywheel effect that accelerates your path to production. Discover the Nimble Flywheel framework for scaling ML.
Infrastructure Production ml infrastructure strategy
Read more →
Machine Learning Project Template with Modern Stack
A modality-aware machine learning project template that gets you from idea to training in minutes using uv, Polars, and LitServe for modern ML workflows.
uv Polars LitServe
Read more →
Claude Code Productivity: I Shipped More, Coded Less
A month with Claude Code: I shipped more projects than ever while paradoxically becoming worse at programming from scratch. Here's what actually happened.
Read more →
A Better ML Training Workflow: Stop Training and Praying
Fix your ML training workflow with the Pre-Flight, In-Flight, Post-Flight framework—catch bugs early, monitor runs, and evaluate beyond vanity metrics.
GPU Training Model Evaluation Data Quality
Read more →
Agentic MLOps: AI Agents Automating Your ML Stack
Explore how agentic MLOps uses AI agents and MCP to autonomously handle data drift, model deployment, and ML pipeline orchestration across your entire stack.
Read more →
A2A Protocol: A Builder's Guide to Agent2Agent
Learn how the A2A Protocol enables secure AI agent communication. Covers Google's open standard, MCP differences, authentication, monetization, orchestration, and Python examples.
Read more →
Tech Trends Next Decade: A 6-Shift Framework
Six shifts define tech trends next decade: Autonomous Economy, Radical Abstraction, Composability, Ecosystem Dynamics, Ambient Intelligence, Next-Gen UX.
IoT Ambient Intelligence Product Strategy
Read more →
AI Agent Frameworks: A Fragmented Landscape
A practical comparison of AI agent frameworks — OpenAI Agents SDK, Google ADK, AWS Bedrock, and Smol Agents — covering architecture, multi-agent systems, and emerging trends.
OpenAI Google ADK AWS Bedrock
Read more →
ML Experiment Tracking as First-Class Infrastructure
Treating ML experiments as disposable runs undermines reproducibility and model quality. Learn why experiment tracking belongs in MLOps infrastructure.
Read more →
Machine Learning Training Infrastructure Guide
Scale ML models faster with the right training infrastructure. Covers distributed training, GPU clouds, spot instances, and cloud-agnostic setups.
Skypilot Distributed Training GPU Cloud
Read more →
MLOps for Startups: A Decision Framework
A decision framework for when to invest in MLOps at your startup — from early-stage flexibility through scaling infrastructure at each growth stage.
Model Deployment Data Pipelines Observability
Read more →
Building Tech Products for India: A First-Principles Playbook
Building tech products for India from first principles—local habits, scale, price sensitivity, and cultural diversity require a full ground-up rethink.
India Product Localization Digital India
Read more →
Rise of the Full Stack Machine Learning Engineer
How the histories of full stack web development and machine learning converged to create the full stack machine learning engineer role.
Full Stack Cloud Infrastructure
Read more →
ML Ecosystem Fragmentation (It Ain't JavaScript)
The ML ecosystem mirrors JavaScript's 2015 framework chaos. I map the fragmentation layers and share how I tackled it by building torchwebio.
PyTorch Ecosystem Fragmentation
Read more →
Pipenv vs Pyenv: Python Version Management
Compare pipenv vs pyenv and learn how combining both tools solves Python version management and dependency isolation in one workflow.
Pipenv Pyenv
Read more →