Framework · 2023–now · AI Present

LLM orchestration frameworks

Chains, tools, memory, and agents as a framework — LangChain-class glue sold as architecture. By 2026: LangChain-fatigue and thinner SDKs.

LLM orchestration frameworks faded as defaults when the abstraction was heavier than the product. The market consolidated into first-party tool APIs and thin SDKs. What sticks is structured tool use and evals; what faded is seventeen wrappers around the same chat completion.

ai framework agents

Cost of the fad

Teams adopted chain/agent frameworks before they had a single reliable tool call. Abstraction layers multiplied while prompt quality stayed flat.

Case studies

AI codebase rewrite pilots that shipped nothing — Six-figure LLM API and SI costs; production cutover: zero

Patterns

AI replaces engineering Resume-driven architecture

Context

Demos consolidate; evals remain

AI pair programming changed how code is typed faster than how it is reviewed. Agent frameworks and standalone vector stores sorted into demos versus durable plumbing; mid-market RAG folded back into Postgres. The permanent layer is familiar: evals that gate deploys, model routing for cost, tool protocols instead of plugin snowflakes, and humans who own production. Autopilot rewrites and vibe-shipped auth middleware are still big-bang migrations with better slides — and a longer on-call.

Compare with

vs Multi-agent orchestration hype vs LLM app frameworks vs RAG as default architecture vs Agent ops / LLM observability

Framework · 2023–2025

Multi-agent orchestration hype

Costly fad

LangGraph, CrewAI, AutoGPT cosplay — agents delegating to agents for tasks one function call could do. Hype peaked; backlash stuck.

$ 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.

Framework · 2023–now

LLM app frameworks

Mutated

LangChain-class glue gave way to vendor SDKs and thin wrappers. The durable pieces are boring: evals, retrieval, and product UX.

Practice · 2023–now

RAG as default architecture

Mutated

Retrieval-augmented generation as the answer to every knowledge problem — then "just stuff the window" as the counter-fad. 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.

Practice · 2023–now

Agent ops / LLM observability

Stuck

Tracing, cost caps, and prompt versioning for production LLM features — mostly constrained tool loops, not autonomous agents. Datadog for tokens.

← Timeline Open graph Costliest fads