# Hanzo — chat, or build with dev > Ask a question, or build what you describe. The whole cloud is one binary and one API — run it here, on your own Kubernetes, or on your laptop. Open source. [Models](https://hanzo.ai/models)/University hanzo.university · Open Enrollment · 25% Usage Credit Rebate # Engineering with the weight of consequence. Move past fragile prompts and toy notebook tutorials. Learn to build, deploy, observe, and budget production-grade multi-agent systems on the Hanzo Cloud Fabric. Every course includes 25% rounded up in usage credits ($25–$125) to fund your cloud labs. [Explore Courses & Credentials](#curriculum)[Read Kai Decision Model Guide](https://hanzo.ai/kai/guide) 25% Usage Credit Rebate (Rounded Up)Every enrollment deposits 25% of tuition directly into your Hanzo Cloud account to fund GPU sandboxes and API calls. 75% Cash Margin · Zero Lab Cost Pedagogical Standard ## Why engineers choose Hanzo University Standard AI courses test multiple-choice memorization. Hanzo University validates production competence through automated CI grading, live deployed sandboxes, and financial governance. Standard of RigorAll coursework is code-evaluated. Submit genuine pull requests, solve SWE-bench issues, and pass automated unit/integration test suites to earn credentials. Financial CeilingsBuild agent loops governed by strict integer micro-USD caps. Pre-call quoting, settlement, and graceful handling of in-band budget refusals. Cognitive DivisionGenerate text only when generation is required. Route with Enso, execute with Zen, and resolve bounded control decisions via Kai with zero output tokens. Zero-Copy ProtocolsGraduate beyond high-latency JSON-RPC. Architect swarms over ZAP (Cap’n Proto RPC) for 10–100x lower overhead and post-quantum encryption. SWE-bench GradingCapstone projects are evaluated by automated test runners in isolated gVisor pods. Passing requires verified code commits, low drift, and valid telemetry. Verifiable CredentialsCertificates are issued as W3C Verifiable Credentials signed by Hanzo Trust with cryptographic proof on-chain and dynamic GitHub SVG badges. Course Catalog ## Courses & Professional Certifications Select an engineering track to inspect syllabi, lab architectures, and enrollment details. All Tracks Agentic Coding Agentic Marketing Reinforcement Learning Systems & Foundation Architecture ENG-401HACE Agentic Coding Systems with Hanzo Dev & Zen 6Architect autonomous coding agents capable of multi-file refactoring, test-driven debugging, and opening verified pull requests using Zen 6, ZAP RPC, and isolated gVisor sandboxes. Core Competencies:Zen 6 (27.3B) & Zen 6 Flash local serving SWE-bench self-healing test execution loops ZAP Zero-Copy Cap’n Proto multi-agent swarms Physical sandbox pod leasing & gVisor isolation $199USD +$50 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) MKT-301HAME Agentic Marketing & Autonomous CampaignsEngineer programmatic growth machines. Generate multi-channel copy, studio images, voiceovers, and videos with locked brand voice, direct CMS hooks, and closed-loop CRO reinforcement. Core Competencies:POST /v1/content/generate CMS pipelines Multi-modal asset generation (Zen 3, Wan 2.2) Event-driven campaign automation with /v1/auto Closed-loop A/B testing & LinUCB CRO optimization $149USD +$38 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) RL-501HARLE Native Reinforcement Learning & Post-TrainingImplement native Gymnasium environments on the Hanzo Cloud Fabric. Fine-tune models with Zoo Gym, design multi-objective reward functions, and train learned routing policies. Core Competencies:Gymnasium HanzoAgentEnv MDP implementation Post-training & fine-tuning with Zoo Gym Multi-objective reward modeling (PPO, GRPO, DPO) LinUCB router policy training via /v1/ai/feedback $249USD +$63 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) SYS-201HCAISE Hanzo AI Systems Engineering FoundationThe definitive engineering standard for production AI. Master the 4-surface parity, finite-state Kai decision loops, hybrid retrieval, and strict integer micro-USD budget governance. Core Competencies:Unified 4-surface API (CLI, SDKs, HTTP, MCP) Finite-state Kai decision loops with 0 output tokens Hybrid retrieval (knowledge, graph, search) Integer micro-USD nested budget governance $149USD +$38 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) PRA-101HCAIP AI Practitioner & Tool IntegrationMaster the fundamentals of the unified cloud. Understand model selection across Zen, Enso, and Kai, integrate the 13 unified MCP tools into your editor, and manage cloud spend. Core Competencies:Navigating api.hanzo.ai and credential management Model family taxonomy and pricing contracts 13 unified MCP tools configuration in Cursor/Claude Usage metering and quota monitoring via CLI $99USD +$25 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) ARC-601HCPAIA Production AI Architect MasterclassEnterprise AI architecture at scale. Build self-hosted zero-trust clusters, configure post-quantum RNS tunnels, deploy global BFT agent meshes, and govern enterprise billing. Core Competencies:Zero-trust cluster provisioning with hanzo up Post-quantum cryptography (ML-KEM, ML-DSA) Global agent mesh & BFT consensus with hanzod Multi-tenant IAM, KMS, and enterprise compliance $499USD +$125 Credits Included [View Course Syllabus →](https://hanzo.ai/university/agentic-coding) Runtime Environment ## Interactive cloud sandboxes. Zero configuration. Every enrolled student receives access to browser-based Hanzo Dev sandboxes running isolated gVisor boundaries. Work with Zen models, test multi-agent ZAP swarms, and run SWE-bench evaluation suites with one command. hanzo-sandbox · dev lease rl-env.py # Run automated Hanzo evaluation harness$ hanzo evals run --suite swe-bench --model zen6 --budget 500000[+] Leased ephemeral sandbox dev_8921b (gVisor runtime, 900s ttl)[+] Spawned root agent session s_019482 (budget: 500,000 micro-USD)[+] Executing turn 1: inspect repo diffs and reproduce failure[+] Kai completion check: done=0.34, verified=0.31 → iterate[+] Executing turn 2: apply AST patch to src/engine.rs[+] Kai completion check: done=0.88, verified=0.91 → PASS[✓] Verification test exit code 0. Total spent: 218,400 micro-USD ($0.218)[✓] Grade: 100/100 · W3C Verifiable Credential Signed by Hanzo Trust Frequently Asked Questions ## Everything you need to know How does the 25% credit rebate work?When you enroll in any course (e.g. $199 for Agentic Coding), 25% of your payment rounded up ($50) is immediately credited to your Hanzo Cloud account. You can use these credits across all models (Zen, Enso, Kai), sandboxes, and MCP tools. How are capstone projects graded?Grading is 100% automated and objective. When you submit your repository, an isolated Hanzo grader pod runs synthetic fault injection, SWE-bench problem sets, and test suites, measuring correctness, budget adherence, and telemetry fidelity. What format are the credentials issued in?Graduates receive W3C Verifiable Credentials signed by Hanzo Trust with cryptographic proof on-chain, as well as a dynamic SVG markdown badge for your GitHub profile showing live verified status and test benchmarks. Do I need a GPU to run these courses?No. You can run all exercises either locally on your laptop CPU/Metal using quantized Zen builds (zen6-flash), or execute in Hanzo Cloud using your included usage credits with zero local setup.