AI Present

Autocomplete that actually understands the file. Changing how code is written faster than how it's reviewed.

Regression tests for nondeterministic models. The unglamorous infrastructure that separates demos from products.

Embeddings back in the database you already run. The quiet winner of the vector DB gold rush.

All-in-one JS runtime that actually feels fast. Winning scripts and side projects while production waits for edge cases.

Retrieval-augmented generation as the answer to every knowledge problem. Often right; often Postgres with pgvector would suffice.

$ Teams stood up vector pipelines, chunking strategies, and rerankers before asking if fine-tuning or a SQL query would answer the question. Retrieval infra became the product.

The shakeout from seventeen LLM wrappers to three vendors and one observability bill. Hype cycle entering boring procurement.

Tracing, cost caps, and prompt versioning for production LLM features. Datadog for tokens.

Chains, agents, and prompt wrappers shipping weekly. The durable pieces will be boring: evals, retrieval, and product UX.

Specialized stores for embeddings. Many workloads are folding back into Postgres extensions — after the vendor tour.

Big-bang rewrite, now with a chatbot. Same failure mode; prettier slides.

$ Exec decks promised autonomous migrations "by next quarter." Teams that skipped tests, ownership, and incremental strangler patterns bought expensive demos and fragile diffs.

Erlang semantics with friendly syntax on the BEAM. The language your functional meetup loves and your employer has never heard of.

LangGraph, CrewAI, AutoGPT cosplay — agents delegating to agents for tasks one function call could do.

$ Autonomous agent swarms burned API budgets on coordination loops that a single prompt and a cron job would have handled. Debugging "which agent lied" became the new on-call sport.

Accept AI output on feel, skip the diff, hope tests exist. Fast demos; fragile ownership.

$ Ship-to-prod without reading the diff became a flex until incidents revealed nobody knew which prompt wrote the auth middleware. Review debt arrived in one weekend demo.

Give the model a goal and a credit card. Viral in 2023; quietly shelved when the loop never stopped spending.

$ Open-ended agent loops burned API credits chasing goals nobody specified well enough to finish. Demo videos outpaced production deployments by orders of magnitude.