Merge pull request #60 from anfredette/rename-to-neuralnav

Complete the rename from Compass to NeuralNav

Author: anfredette

.streamlit/config.toml

@@ -4,6 +4,7 @@ gatherUsageStats = false
 [server]
 headless = true
 port = 8501
+enableStaticServing = true
 
 [theme]
 primaryColor = "#1f77b4"

CLAUDE.md

@@ -4,7 +4,7 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 
 ## Project Overview
 
-This repository contains the architecture design for **Compass**, an open-source system that guides users from concept to production-ready LLM deployments through a conversational AI and intelligent capacity planning.
+This repository contains the architecture design for **NeuralNav**, an open-source system that guides users from concept to production-ready LLM deployments through a conversational AI and intelligent capacity planning.
 
 **Key Principle**: This is a **Phase 1 POC implementation** based on the architecture design. The core functionality is complete and working end-to-end.
 
@@ -82,7 +82,7 @@ The system translates high-level user intent into technical specifications:
 
 ### Architecture Overview
 
-Compass is structured as a layered architecture:
+NeuralNav is structured as a layered architecture:
 
 **UI Layer** (Horizontal - Presentation):
 - **Conversational Interface, Specification Editor, Recommendation Visualizer, Monitoring Dashboard**
@@ -184,7 +184,7 @@ The recommendation engine uses **multi-criteria scoring** to rank configurations
 
 ### Terminology Consistency
 
-- Use "**Compass**" as the project name
+- Use "**NeuralNav**" as the project name
 - Use "**TTFT**" for Time to First Token (not "time-to-first-token")
 - Use "**ITL**" for Inter-Token Latency (Phase 2 terminology, replaces TPOT)
 - Use "**SLO**" for Service Level Objective

CONTRIBUTING.md

@@ -1,6 +1,6 @@
-# Contributing to Compass
+# Contributing to NeuralNav
 
-Thank you for your interest in contributing to Compass! This document provides guidelines for contributing to the project.
+Thank you for your interest in contributing to NeuralNav! This document provides guidelines for contributing to the project.
 
 ## Table of Contents
 
@@ -26,24 +26,24 @@ This section describes the complete workflow from setup to submitting a pull req
 
 **2. Clone your fork locally:**
 ```bash
-git clone https://github.com/YOUR_USERNAME/compass.git
-cd compass
+git clone https://github.com/YOUR_USERNAME/neuralnav.git
+cd neuralnav
 ```
 
 **3. Set up remotes:**
 ```bash
 # Your fork is already set as 'origin' by the clone
 # Add the main repository as 'upstream'
-git remote add upstream https://github.com/redhat-et/compass.git
+git remote add upstream https://github.com/redhat-et/neuralnav.git
 ```
 
 Verify your remotes:
 ```bash
 git remote -v
-# origin    https://github.com/YOUR_USERNAME/compass.git (fetch)
-# origin    https://github.com/YOUR_USERNAME/compass.git (push)
-# upstream  https://github.com/redhat-et/compass.git (fetch)
-# upstream  https://github.com/redhat-et/compass.git (push)
+# origin    https://github.com/YOUR_USERNAME/neuralnav.git (fetch)
+# origin    https://github.com/YOUR_USERNAME/neuralnav.git (push)
+# upstream  https://github.com/redhat-et/neuralnav.git (fetch)
+# upstream  https://github.com/redhat-et/neuralnav.git (push)
 ```
 
 **4. Set up your development environment** by following the [Quick Start guide](README.md#quick-start).
@@ -334,7 +334,7 @@ make format
 
 ## License
 
-By contributing to Compass, you agree that your contributions will be licensed under the same license as the project (see [LICENSE](LICENSE)).
+By contributing to NeuralNav, you agree that your contributions will be licensed under the same license as the project (see [LICENSE](LICENSE)).
 
 ## Questions?
 
@@ -343,4 +343,4 @@ If you have questions not covered here, please:
 2. Open a new issue with the "question" label
 3. Reach out to maintainers
 
-Thank you for contributing to Compass!
+Thank you for contributing to NeuralNav!

DOCKER.md

@@ -1,10 +1,10 @@
-# Docker Deployment Guide for Compass
+# Docker Deployment Guide for NeuralNav
 
-This guide explains how to run Compass using Docker and Docker Compose.
+This guide explains how to run NeuralNav using Docker and Docker Compose.
 
 ## Overview
 
-Compass is containerized into the following services:
+NeuralNav is containerized into the following services:
 
 - **postgres**: PostgreSQL database for benchmarks and deployment outcomes
 - **ollama**: Ollama LLM service for intent extraction (qwen2.5:7b model)
@@ -70,7 +70,7 @@ This repository includes a **Makefile** that provides convenient shortcuts for a
 | **Clean everything** | `docker-compose down -v --remove-orphans` | `make docker-clean` |
 | **Backend shell** | `docker-compose exec backend /bin/bash` | `make docker-shell-backend` |
 | **UI shell** | `docker-compose exec ui /bin/bash` | `make docker-shell-ui` |
-| **PostgreSQL shell** | `docker-compose exec postgres psql -U compass -d compass` | `make docker-shell-postgres` |
+| **PostgreSQL shell** | `docker-compose exec postgres psql -U neuralnav -d neuralnav` | `make docker-shell-postgres` |
 
 ### Examples
 
@@ -179,20 +179,20 @@ make docker-ps
 
 ```bash
 # Access PostgreSQL CLI
-docker-compose exec postgres psql -U compass -d compass
+docker-compose exec postgres psql -U neuralnav -d neuralnav
 make docker-shell-postgres
 
 # Run database migrations/scripts
-docker-compose exec postgres psql -U compass -d compass -f /scripts/schema.sql
+docker-compose exec postgres psql -U neuralnav -d neuralnav -f /scripts/schema.sql
 
 # Load benchmark data
 docker-compose exec backend python scripts/load_benchmarks.py
 
 # Backup database
-docker-compose exec postgres pg_dump -U compass compass > backup.sql
+docker-compose exec postgres pg_dump -U neuralnav neuralnav > backup.sql
 
 # Restore database
-docker-compose exec -T postgres psql -U compass compass < backup.sql
+docker-compose exec -T postgres psql -U neuralnav neuralnav < backup.sql
 ```
 
 ### Ollama Operations
@@ -216,8 +216,8 @@ Create a `.env` file in the project root to customize configuration:
 
 ```env
 # Database Configuration
-POSTGRES_DB=compass
-POSTGRES_USER=compass
+POSTGRES_DB=neuralnav
+POSTGRES_USER=neuralnav
 POSTGRES_PASSWORD=your_secure_password
 
 # Ollama Configuration
@@ -324,11 +324,11 @@ curl http://localhost:11434/  # Ollama
 
 ### Network Topology
 
-All services communicate via a bridge network (`compass-network`):
+All services communicate via a bridge network (`neuralnav-network`):
 
 ```
 ┌─────────────────────────────────────────────┐
-│           compass-network (bridge)          │
+│           neuralnav-network (bridge)          │
 │                                             │
 │  ┌─────────┐    ┌─────────┐    ┌────────┐   │
 │  │   UI    │───▶│ Backend │───▶│Postgres│   │
@@ -382,7 +382,7 @@ docker-compose logs postgres  # Use docker-compose for PostgreSQL
 docker-compose exec ollama ollama list
 
 # Verify database schema
-docker-compose exec postgres psql -U compass -c "\dt"
+docker-compose exec postgres psql -U neuralnav -c "\dt"
 ```
 
 ### Database Connection Errors
@@ -397,7 +397,7 @@ docker-compose logs postgres
 
 **Test database connection:**
 ```bash
-docker-compose exec postgres pg_isready -U compass
+docker-compose exec postgres pg_isready -U neuralnav
 ```
 
 **Reset database (WARNING: deletes all data):**
@@ -474,7 +474,7 @@ Example nginx configuration:
 ```nginx
 server {
     listen 80;
-    server_name compass.example.com;
+    server_name neuralnav.example.com;
 
     location / {
         proxy_pass http://localhost:8501;
@@ -513,6 +513,6 @@ server {
 ## Support
 
 For issues and questions:
-- Check [GitHub Issues](https://github.com/yourusername/compass/issues)
+- Check [GitHub Issues](https://github.com/yourusername/neuralnav/issues)
 - Review Docker logs: `docker-compose logs -f`
 - Verify service health: `docker-compose ps`

LICENSE

@@ -186,7 +186,7 @@
       same "printed page" as the copyright notice for easier
       identification within third-party archives.
 
-   Copyright 2025 The authors of Compass
+   Copyright 2025 The authors of NeuralNav
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

Makefile

@@ -1,6 +1,6 @@
-# Compass - Makefile
+# NeuralNav - Makefile
 #
-# This Makefile provides common development tasks for Compass.
+# This Makefile provides common development tasks for NeuralNav.
 # Supports macOS and Linux.
 
 .PHONY: help
@@ -40,13 +40,13 @@ CONTAINER_TOOL ?= $(shell \
 
 # Configuration
 REGISTRY ?= quay.io
-REGISTRY_ORG ?= vllm-assistant
+REGISTRY_ORG ?= neuralnav
 SIMULATOR_IMAGE ?= vllm-simulator
 SIMULATOR_TAG ?= latest
 SIMULATOR_FULL_IMAGE := $(REGISTRY)/$(REGISTRY_ORG)/$(SIMULATOR_IMAGE):$(SIMULATOR_TAG)
 
 OLLAMA_MODEL ?= qwen2.5:7b
-KIND_CLUSTER_NAME ?= compass-poc
+KIND_CLUSTER_NAME ?= neuralnav
 
 PGDUMP_INPUT ?= data/integ-oct-29.sql
 PGDUMP_OUTPUT ?= data/benchmarks_GuideLLM.json
@@ -213,15 +213,15 @@ stop: ## Stop all services
 		kill $$(cat $(BACKEND_PID)) 2>/dev/null || true; \
 		rm -f $(BACKEND_PID); \
 	fi
-	@# Kill any remaining Compass processes by pattern matching
+	@# Kill any remaining NeuralNav processes by pattern matching
 	@pkill -f "streamlit run ui/app.py" 2>/dev/null || true
 	@pkill -f "uvicorn src.api.routes:app" 2>/dev/null || true
 	@# Give processes time to exit gracefully
 	@sleep 1
 	@# Force kill if still running
 	@pkill -9 -f "streamlit run ui/app.py" 2>/dev/null || true
 	@pkill -9 -f "uvicorn src.api.routes:app" 2>/dev/null || true
-	@printf "$(GREEN)✓ All Compass services stopped$(NC)\n"
+	@printf "$(GREEN)✓ All NeuralNav services stopped$(NC)\n"
 	@# Don't stop Ollama as it might be used by other apps
 	@printf "$(YELLOW)Note: Ollama left running (use 'pkill ollama' to stop manually)$(NC)\n"
 
@@ -255,7 +255,7 @@ open-backend: ## Open backend API docs in browser
 
 build-simulator: ## Build vLLM simulator Docker image
 	@printf "$(BLUE)Building simulator image...$(NC)\n"
-	cd $(SIMULATOR_DIR) && $(CONTAINER_TOOL) build -t vllm-simulator:latest -t $(SIMULATOR_FULL_IMAGE) .
+	$(CONTAINER_TOOL) build -f $(SIMULATOR_DIR)/Dockerfile -t vllm-simulator:latest -t $(SIMULATOR_FULL_IMAGE) .
 	@printf "$(GREEN)✓ Simulator image built:$(NC)\n"
 	@printf "  - vllm-simulator:latest\n"
 	@printf "  - $(SIMULATOR_FULL_IMAGE)\n"
@@ -347,7 +347,7 @@ docker-shell-ui: ## Open shell in UI container
 	@docker-compose exec ui /bin/bash
 
 docker-shell-postgres: ## Open PostgreSQL shell in container
-	@docker-compose exec postgres psql -U compass -d compass
+	@docker-compose exec postgres psql -U neuralnav -d neuralnav
 
 ##@ Kubernetes Cluster
 
@@ -383,38 +383,38 @@ clean-deployments: ## Delete all InferenceServices from cluster
 
 db-start: ## Start PostgreSQL container for benchmark data
 	@printf "$(BLUE)Starting PostgreSQL...$(NC)\n"
-	@if $(CONTAINER_TOOL) ps -a --format '{{.Names}}' | grep -q '^compass-postgres$$'; then \
-		if $(CONTAINER_TOOL) ps --format '{{.Names}}' | grep -q '^compass-postgres$$'; then \
+	@if $(CONTAINER_TOOL) ps -a --format '{{.Names}}' | grep -q '^neuralnav-postgres$$'; then \
+		if $(CONTAINER_TOOL) ps --format '{{.Names}}' | grep -q '^neuralnav-postgres$$'; then \
 			printf "$(YELLOW)PostgreSQL already running$(NC)\n"; \
 		else \
-			$(CONTAINER_TOOL) start compass-postgres; \
+			$(CONTAINER_TOOL) start neuralnav-postgres; \
 			printf "$(GREEN)✓ PostgreSQL started$(NC)\n"; \
 		fi \
 	else \
-		$(CONTAINER_TOOL) run --name compass-postgres -d \
-			-e POSTGRES_PASSWORD=compass \
-			-e POSTGRES_DB=compass \
+		$(CONTAINER_TOOL) run --name neuralnav-postgres -d \
+			-e POSTGRES_PASSWORD=neuralnav \
+			-e POSTGRES_DB=neuralnav \
 			-p 5432:5432 \
 			postgres:16; \
 		sleep 3; \
 		printf "$(GREEN)✓ PostgreSQL started on port 5432$(NC)\n"; \
 	fi
-	@printf "$(BLUE)Database URL:$(NC) postgresql://postgres:compass@localhost:5432/compass\n"
+	@printf "$(BLUE)Database URL:$(NC) postgresql://postgres:neuralnav@localhost:5432/neuralnav\n"
 
 db-stop: ## Stop PostgreSQL container
 	@printf "$(BLUE)Stopping PostgreSQL...$(NC)\n"
-	@$(CONTAINER_TOOL) stop compass-postgres 2>/dev/null || true
+	@$(CONTAINER_TOOL) stop neuralnav-postgres 2>/dev/null || true
 	@printf "$(GREEN)✓ PostgreSQL stopped$(NC)\n"
 
 db-remove: db-stop ## Stop and remove PostgreSQL container
 	@printf "$(BLUE)Removing PostgreSQL container...$(NC)\n"
-	@$(CONTAINER_TOOL) rm compass-postgres 2>/dev/null || true
+	@$(CONTAINER_TOOL) rm neuralnav-postgres 2>/dev/null || true
 	@printf "$(GREEN)✓ PostgreSQL container removed$(NC)\n"
 
 db-init: db-start ## Initialize PostgreSQL schema
 	@printf "$(BLUE)Initializing PostgreSQL schema...$(NC)\n"
 	@sleep 2
-	@$(CONTAINER_TOOL) exec -i compass-postgres psql -U postgres -d compass < scripts/schema.sql
+	@$(CONTAINER_TOOL) exec -i neuralnav-postgres psql -U postgres -d neuralnav < scripts/schema.sql
 	@printf "$(GREEN)✓ Schema initialized$(NC)\n"
 
 db-load-synthetic: db-start ## Load synthetic benchmark data (appends)
@@ -457,19 +457,19 @@ db-convert-pgdump: db-start ## Convert PostgreSQL dump to JSON format
 	@printf "$(GREEN)✓ Created $(PGDUMP_OUTPUT)$(NC)\n"
 
 db-shell: ## Open PostgreSQL shell
-	@$(CONTAINER_TOOL) exec -it compass-postgres psql -U postgres -d compass
+	@$(CONTAINER_TOOL) exec -it neuralnav-postgres psql -U postgres -d neuralnav
 
 db-query-traffic: ## Query unique traffic patterns from database
 	@printf "$(BLUE)Querying unique traffic patterns...$(NC)\n"
-	@$(CONTAINER_TOOL) exec -i compass-postgres psql -U postgres -d compass -c \
+	@$(CONTAINER_TOOL) exec -i neuralnav-postgres psql -U postgres -d neuralnav -c \
 		"SELECT DISTINCT mean_input_tokens, mean_output_tokens, COUNT(*) as num_benchmarks \
 		FROM exported_summaries \
 		GROUP BY prompt_tokens, output_tokens \
 		ORDER BY prompt_tokens, output_tokens;"
 
 db-query-models: ## Query available models in database
 	@printf "$(BLUE)Querying available models...$(NC)\n"
-	@$(CONTAINER_TOOL) exec -i compass-postgres psql -U postgres -d compass -c \
+	@$(CONTAINER_TOOL) exec -i neuralnav-postgres psql -U postgres -d neuralnav -c \
 		"SELECT DISTINCT model_hf_repo, hardware, hardware_count, COUNT(*) as num_benchmarks \
 		FROM exported_summaries \
 		GROUP BY model_hf_repo, hardware, hardware_count \

README.md

@@ -1,4 +1,4 @@
-# <img src="docs/compass-logo.ico" alt="Compass" width="32" style="vertical-align: middle;"/> Compass
+# <img src="docs/neuralnav-logo.ico" alt="NeuralNav" width="32" style="vertical-align: middle;"/> NeuralNav
 
 **Confidently navigate LLM deployments from concept to production.**
 
@@ -8,9 +8,9 @@
 
 The system addresses a critical challenge: **how do you translate business requirements into the right model and infrastructure choices without expensive trial-and-error?**
 
-Compass guides you from concept to production LLM deployments through SLO-driven capacity planning. Conversationally define your requirements—Compass translates them into traffic profiles, performance targets, and cost constraints. Get intelligent model and GPU recommendations based on real benchmarks. Explore alternatives, compare tradeoffs, deploy with one click, and monitor actual performance—staying on course as your needs evolve.
+NeuralNav guides you from concept to production LLM deployments through SLO-driven capacity planning. Conversationally define your requirements—NeuralNav translates them into traffic profiles, performance targets, and cost constraints. Get intelligent model and GPU recommendations based on real benchmarks. Explore alternatives, compare tradeoffs, deploy with one click, and monitor actual performance—staying on course as your needs evolve.
 
-The code in this repository implements the **Compass Phase 2 MVP** with production-grade data management. Phase 1 (POC) demonstrated the end-to-end workflow with synthetic data. Phase 2 adds PostgreSQL for benchmark storage, a traffic profile framework aligned with GuideLLM standards, experience-driven SLO mapping, and p95 percentile targets for conservative guarantees.
+The code in this repository implements the **NeuralNav Phase 2 MVP** with production-grade data management. Phase 1 (POC) demonstrated the end-to-end workflow with synthetic data. Phase 2 adds PostgreSQL for benchmark storage, a traffic profile framework aligned with GuideLLM standards, experience-driven SLO mapping, and p95 percentile targets for conservative guarantees.
 
 ### Key Features
 
@@ -59,15 +59,15 @@ make cluster-start  # Optional: Create local KIND cluster with vLLM simulator fo
 
 Then open [http://localhost:8501](http://localhost:8501) in your browser.
 
-**Note**: PostgreSQL runs as a Docker container (`compass-postgres`) with benchmark data. All `db-load-*` commands append to existing data. Use `make db-reset` first for a clean database. 
+**Note**: PostgreSQL runs as a Docker container (`neuralnav-postgres`) with benchmark data. All `db-load-*` commands append to existing data. Use `make db-reset` first for a clean database. 
 
 **Stop everything:**
 ```bash
 make stop           # Stop services
 make cluster-stop   # Delete cluster (optional)
 ```
 
-### Using Compass
+### Using NeuralNav
 
 1. **Describe your use case** in the chat interface
    - Example: "I need a customer service chatbot for 5000 users with low latency"
@@ -93,7 +93,7 @@ The POC includes 3 pre-configured scenarios (see [data/demo_scenarios.json](data
 
 ## Architecture Highlights
 
-Compass implements an **8-component architecture** with:
+NeuralNav implements an **8-component architecture** with:
 
 - **Conversational Interface** (Streamlit) - Chat-based requirement gathering with interactive exploration
 - **Context & Intent Engine** - LLM-powered extraction of deployment specs
@@ -169,7 +169,7 @@ make clean                   # Remove generated files
 
 ## vLLM Simulator Mode
 
-Compass includes a **GPU-free simulator** for local development:
+NeuralNav includes a **GPU-free simulator** for local development:
 
 - **No GPU required** - Run deployments on any laptop
 - **OpenAI-compatible API** - `/v1/completions` and `/v1/chat/completions`