vicentereig

Your single-tool approach is a solid starting point. As it grows, you might hit context window limits and find the prompt getting unwieldy. Things like why is this prompt choking on 1.5MB of JSON coming from this other API/Tool?

When you look at systems like Codex CLI, they run at least four separate LLM subsystems: (1) the main agent prompt, (2) a summarizer model that watches the reasoning trace and produces user-facing updates like "Searching for test files...", (3) compaction and (4) a reviewer agent. Each one only sees the context it needs. Like a function with their inputs and outputs. Total tokens stay similar, but signal density per prompt goes up.

DSPy.rb[0] enables this pattern in Ruby: define typed Signatures for each concern, compose them as Modules/Prompting Techniques (simple predictor, CoT, ReAct, CodeAct, your own, ...), and let each maintain its own memory scope. Three articles that show this:

- "Ephemeral Memory Chat"[1] — the Two-Struct pattern (rich storage vs. lean prompt context) plus cost-based routing between cheap and expensive models.

- "Evaluator Loops"[2] — decompose generation from evaluation: a cheap model drafts, a smarter model critiques, each with its own focused signature.

- "Workflow Router"[3] — route requests to the right model based on complexity, only escalate to expensive LLMs when needed.

And since you're already using RubyLLM, the dspy-ruby_llm adapter lets you keep your provider setup while gaining the decomposition benefits.

Thanks for coming to my TED talk. Let me know if you need someone to bounce ideas off.

RubyLLM gives you a clean API for LLM calls and tool definitions. You're still writing prompts and managing conversations directly.

DSPy.rb treats prompts as functions with typed signatures. You define inputs/outputs and the framework handles prompt construction, JSON parsing, and structured extraction. Two articles that might help:

1. "Building Your First ReAct Agent" shows how to build tool-using agents with type-safe tool definitions [0].

2. "Building Chat Agents with Ephemeral Memory" demonstrates context engineering patterns (what the LLM sees vs. what you store), cost-based routing between models, and memory management [1].

The article's approach (RubyLLM + single tool) works great for simple cases. DSPy.rb shines when you need to decompose into multiple specialized modules with different concerns. Some examples: separate signatures for classification vs. response generation, each optimized independently with separate context windows and memory to maintain.

Would love to learn how dspy.rb is working for you!

Note that RubyLLM and DSPy.rb aren't mutually exclusive (`gem 'dspy-ruby_llm'`) adapter gives us access to a TON of providers.

I’ve been using this pattern for API clients[0] and CLIs[1]: define the shape once with Sorbet, get automatic JSON Schema generation when you need it.

For generating (with LLMs) API clients and CLIs it’s especially useful—define the shape once, get validation at ingress/egress for free.

DSPy is a programming paradigm. I like to look at it like the MVC for the Web. You define Signatures[0]: typed contracts governing the relationship between your models and your app. Signatures model prompts as functions too, without leaving Ruby. Then compose them into modules (Predict, ChainOfThought, ReAct, your own). The framework can automatically optimize prompts based on your metrics.

DSPy.rb brings the DSPy paradigm's tooling (optimizers, evaluation loops) to Ruby. Comes with OpenTelemetry OOTB. It also borrows BAML's schema format for 85% token savings vs JSON Schema in complex signatures. [1]

Everyone is talking about prompt, context, and harness engineering -and I agree they are good ways to frame how to build workflows and agents- this is just programming really.

I'm working on DSPy.rb [1] and this could pair nicely for multi-step reasoning chains.

Curious - any plans for async gem support?

It'd be helpful if you could provide a couple of examples of that _boilerplate content_ looks like.

I led the team migrating to Contentful (CF) a hefty documentation website [1] from a homegrown CMS based on MDX. 600+ mdx articles, multiple GBs of media. This was about a year ago, so my knowledge might be outdated.

What worked for us was focusing our workflow as if CF were a database. I personally borrowed a bunch of concepts of how Rails' Active Record allows you, or your CI pipeline, to manage your database schema. Things like:

- have CF doing some gymnastics to keep track of migrations ran.

- supporting ephemeral CF environments so each developer could work without stepping on each others toes.

- promoting environments on PR merges

Being able to track changes in CF's schema through migrations, allowed us to build other useful stuff like:

- structuring articles in a way that the navigation under each section is not a pain to rebuild.

This might be what you mean when you talk about _boilerplate content_. Didn't find anything at the time, but it's possible that someone else has built a framework of sorts.

It took us a few weeks to prepare ours before working on actually migrating content. However, it supported our work along the way during the incremental and continuous migration we planned.

And then we were in a good spot to actually start migrating content. Transpiling the MDX to the JSON CF can ingest, uploading media assets, rewiring the UI (nextjs) to fetch contents from CF instead of the local MDX source.

Anyway, let me know if you want to bounce some ideas off. Contact details in my profile.