AI Engineering

Posts on AI engineering — the discipline of building real products on top of LLMs. Practical writing on RAG, agents, prompt engineering, vector databases, evaluations, and the production realities of shipping AI features that don’t fall apart in week three.

Giving AI Agents Memory in 2026 — Mem0, Zep, and the Patterns That Work

Why agents need memory beyond the context window, the 2026 tools (Mem0, Zep, custom layers), summary vs episodic memory, retrieval, and the patterns from production agents.

Sandboxed Code Execution for AI Agents — E2B, Modal, Daytona, and the 2026 Stack

Why agents need sandboxed code execution, the 2026 platforms (E2B, Modal, Daytona, Fly Machines, custom microVMs), tradeoffs, and how to wire it into an agent.

AI Coding Assistants ROI in 2026 — The Honest Numbers

What AI coding assistants actually deliver in 2026. Where they save hours, where they create new work, the productivity research, and the adoption patterns of teams that ship faster vs teams that hit dead ends.

1M-Token Context Windows in 2026 — When They Help, When They Hurt

How to actually use 1M-token context windows. The ‘just put it all in context’ temptation, when it works, when RAG still wins, prompt caching, and cost.

Agentic RAG in 2026 — When Retrieval Becomes a Tool, Not a Pipeline

Why agentic RAG often beats one-shot RAG. Tool-based retrieval, decomposition, query rewriting, self-reflection, citations, and the production patterns that ship in 2026.

LLM Security in 2026 — Prompt Injection, Data Exfiltration, and Defense in Depth

LLM security threats and defenses in 2026. Direct + indirect prompt injection, exfiltration via tool calls or markdown, jailbreaks, and the layered defenses (input tagging, output filtering, allow-lists, OPA, sandboxing).

LLM Observability in 2026 — LangSmith, Langfuse, Helicone, and OpenTelemetry

What to track in LLM apps, the tooling landscape (LangSmith, Langfuse, Helicone, Phoenix), the OTel GenAI conventions, and the metrics-and-traces playbook for production AI.

Rerankers in RAG — The Underrated Quality Multiplier in 2026

Rerankers turn ‘pretty good RAG’ into ‘great RAG’ for one extra API call. Cross-encoders explained, Cohere Rerank vs BGE-Reranker vs Jina, two-stage retrieval architecture, latency, cost, and implementation.

Embedding Models in 2026 — OpenAI, Voyage, Cohere, BGE, and How to Pick

A practical 2026 guide to picking an embedding model. OpenAI text-embedding-3 vs Voyage vs Cohere vs open BGE / Nomic. Quality on MTEB, cost, dimensions, multilingual, and how to evaluate on your own data.

Voice Agents and Realtime LLM APIs in 2026 — How They Actually Work

A practical look at building voice agents in 2026. Realtime LLM APIs (OpenAI Realtime, Anthropic, Gemini Live), end-to-end latency, ASR and TTS, interruption handling, and the production patterns from real deployments.