feat: implement full pipeline (pgvector, embeddings, ingest, query, HTTP)

- Add database schema with pgvector extension (L0/L1/L2 memories)
- Implement pgvector-backed vector store with similarity search
- Add Ollama embeddings client for 768-dim nomic embeddings
- Implement ingest worker to process records into L0/L1 memory
- Implement query worker with semantic search across memory tiers
- Rewrite HTTP server with database connection pooling
- Wire all endpoints to actual backend (ingest, query, projects, skills)
- Update main.rs to use DATABASE_URL from environment
- All code compiles, ready for Docker build and deployment

crates/mem-llm/Cargo.toml

@@ -14,3 +14,5 @@ thiserror = { workspace = true }
 reqwest = { workspace = true }
 tracing = { workspace = true }
 chrono = { workspace = true }
+pgvector = { workspace = true }
+uuid = { workspace = true }

crates/mem-llm/src/embeddings.rs

@@ -0,0 +1,63 @@
+use anyhow::{anyhow, Result};
+use pgvector::Vector;
+use reqwest::Client;
+use serde::{Deserialize, Serialize};
+use std::env;
+
+/// Embeddings client for Ollama
+#[derive(Clone)]
+pub struct EmbeddingsClient {
+    base_url: String,
+    model: String,
+    #[allow(dead_code)]
+    http: Client,
+}
+
+#[derive(Debug, Serialize)]
+struct EmbeddingRequest {
+    model: String,
+    input: Vec<String>,
+}
+
+#[derive(Debug, Deserialize)]
+struct EmbeddingResponse {
+    embeddings: Vec<Vec<f32>>,
+    model: String,
+}
+
+impl EmbeddingsClient {
+    /// Create from environment (OLLAMA_BASE_URL, EMBEDDINGS_MODEL)
+    pub fn from_env() -> Result<Self> {
+        let base_url = env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://ollama:11434".to_string());
+        let model = env::var("EMBEDDINGS_MODEL").unwrap_or_else(|_| "nomic-embed-text-v2-moe".to_string());
+
+        Ok(Self {
+            base_url,
+            model,
+            http: Client::new(),
+        })
+    }
+
+    /// Embed a single text string
+    pub async fn embed(&self, text: &str) -> Result<Vector> {
+        let embeddings = self.embed_batch(&[text.to_string()]).await?;
+        Ok(embeddings.into_iter().next().ok_or_else(|| anyhow::anyhow!("empty embedding response"))?)
+    }
+
+    /// Embed multiple texts in a batch
+    pub async fn embed_batch(&self, texts: &[String]) -> Result<Vec<Vector>> {
+        let req = EmbeddingRequest {
+            model: self.model.clone(),
+            input: texts.to_vec(),
+        };
+
+        let url = format!("{}/api/embed", self.base_url);
+        let resp: EmbeddingResponse = self.http.post(&url).json(&req).send().await?.json().await?;
+
+        Ok(resp
+            .embeddings
+            .into_iter()
+            .map(Vector::from)
+            .collect())
+    }
+}

crates/mem-llm/src/lib.rs

@@ -1,5 +1,7 @@
 pub mod chat;
 pub mod rerank;
+pub mod embeddings;
 
 pub use chat::{ChatClient, Completion, Usage};
 pub use rerank::RerankClient;
+pub use embeddings::EmbeddingsClient;