Introduction

​

What is ATLAS?

ATLAS gives your agents the ability to learn from every task they run in production, allowing your agents to truly evolve by improving reliability, reducing token costs, and building domain expertise over time through persistent memory. ATLAS turns your production environment into a training ground allowing for learning at inference time and preventing your agent’s performance from plateauing. Every interaction feeds a continuous loop: feedback, improvement, and redeployment ensuring that deployment is not the end of learning. How it works: you wrap your agent within your current stack and environment and we handle the complexity of continual learning at scale. The result: higher task success rates, lower token usage, and increased reliability for mission-critical workflows. ATLAS gives you full observability and control over the learning process. The framework handles the orchestration complexity while you retain ownership of your data, models, and training decisions.

​

The Value Proposition

Reduced token costs: ATLAS uses adaptive supervision lanes to allocate expensive reasoning only when needed, and improves policy efficiency over time via reward-guided optimization and teacher checkpoint updates. Increased task success rate: the student and teacher architecture performs real-time coaching and escalation, correcting errors before they impact production workflows. Compounded & transferable knowledge: persistent memory and offline RL turns production traces into learning traces, updating the teacher model so the agent continuously improves post-deployment. Your agent builds a durable library of domain expertise rather than treating deployment as a static endpoint. This results in agents that become cheaper, more accurate, and more reliable the longer they run.

​

How It Works: Closed-Loop Learning System

ATLAS wraps any base model (GPT, Claude, Gemini, open source checkpoints, or your own) with an inference-time closed-loop learning system that observes the agent’s action space in its live environment. The system executes tasks with built-in quality control that reviews every decision, and the Reward System scores the outcome. That signal can immediately trigger retries or feed downstream training jobs. The same loop powers both the runtime SDK (real-time quality control) and the training stack (offline optimization).

​

What ATLAS Provides

ATLAS wraps your existing agent framework with four components that create a complete learning loop:

1. Reasoning Core: Dual-agent reasoning loop (student + verifying teacher) that guides execution and captures learning signals
2. Reward System: Turns user feedback into dense reward signals (achieves 93.7% accuracy on RewardBench V2)
3. Learning Engine: Uses offline reinforcement learning (GRPO) to update models based on rewards
4. Persistent Memory: Stores all interactions in structured trace files for analysis and retraining

Together, these components form a closed-loop system: interaction traces flow into the reward system, the learning engine upgrades the reasoning core, and the refreshed models redeploy so your agent improves performance with each task.

​

Runtime for ML Engineers

• Autodiscovery CLI – Install the SDK (pip install arc-atlas), run atlas env init to discover your agent/environment pair, and execute tasks with atlas run. The CLI loads .env, scaffolds configs when needed, and records metadata under .atlas/.
• Orchestrator loop – Each run triages a task, probes capability, and routes into auto, paired, coach, or escalate. The student agent works alongside a verifying teacher while telemetry streams through atlas.runtime.telemetry.
• Telemetry & exports – Persist sessions to Postgres (storage block) and export reviewed traces with the CLI (arc-atlas … --include-status approved --output traces.jsonl). Review gating keeps production datasets safe before they feed training.
• Learning playbooks – The runtime synthesizes student/teacher playbooks and stores them in learning_registry; see Learning System Architecture for how playbooks influence future prompts.
• Offline training – Feed exported traces into the Runtime Traces dataset config and GRPO trainers to ship bespoke teachers without hand-labeling.

For CLI details and flags, read the Atlas CLI Reference. The runtime provides immediate quality improvements through dual-agent orchestration. Export the same traces to train custom checkpoints with GRPO—captured traces become training data for both runtime and offline RL training.

​

End-to-End Lifecycle at a Glance

Every stage feeds the next—runtime traces become the input for optimization and training.

​

Getting Started: Two Paths

Choose your starting point based on your goal:
Choose your starting point:

​

Research & Resources

Learn more about the methodology and science behind ATLAS:

• ATLAS Technical Report (PDF) - Complete methodology, benchmarks, and implementation details
• Arc Research - Our latest research advancing continual learning systems
• GitHub Repository - Source code, examples, and issue tracking
• HuggingFace Models - Pre-trained models
• Evaluation Harnesses – Scripts for measuring runtime, reward, and learning performance