Posts tagged #AI

$ cat can-ai-replace-humans.mdx

Jul 9, 2026Innovation

Can AI Replace Humans?

AI can replace some tasks, but replacing humans is a much bigger question. The real change is not human versus AI — it is people who learn to work with AI versus people who ignore it.

#AI#Future#Work#Opinion

~/blog/designing-llm-evals-that-catch-regressions.mdx

$ cat designing-llm-evals-that-catch-regressions.mdx

#AI#LLM

Jul 9, 2026AI

Designing LLM Evals That Catch Real Regressions

LLM features break in ways normal unit tests miss. A useful eval suite starts with real examples, checks behavior instead of vibes, and separates model quality from product quality.

#AI#LLM#Testing#Evaluation

~/blog/streaming-llm-responses-nextjs.mdx

$ cat streaming-llm-responses-nextjs.mdx

#AI#LLM

Jul 6, 2026AI

Streaming LLM Responses in Next.js Without a Framework

Token-by-token streaming makes an AI feature feel instant. Here's how I stream model output straight from a Next.js Route Handler using the Web Streams API — no extra SDK, no client library, about 40 lines total.

#AI#LLM#Next.js#Streaming

~/blog/rag-pgvector-postgres.mdx

$ cat rag-pgvector-postgres.mdx

RAG

#RAG#Embeddings

Jul 2, 2026AI

RAG on Postgres You Already Have: pgvector From Scratch

You don't need a dedicated vector database to ship retrieval-augmented generation. With the pgvector extension, the Postgres instance behind your app becomes a semantic search engine. Here's the whole pipeline — chunking, embedding, storing, and querying.

#RAG#Embeddings#PostgreSQL#AI

~/blog/structured-outputs-tool-calling-llms.mdx

$ cat structured-outputs-tool-calling-llms.mdx

LLM

#LLM#Tool Calling

Jun 28, 2026AI

Getting Reliable JSON Out of an LLM

"Respond in JSON" in the prompt is not a contract — it's a suggestion the model breaks under pressure. Here's how I get schema-valid structured data every time using tool calling, schema validation, and a tight repair loop.

#LLM#Tool Calling#AI#TypeScript