Home › Tracks › Agentic Systems & Harness Engineering
Elite track
Orchestrate deterministic fleets of cloud-native agents with rigorous outer-loop verification.
Master advanced harness engineering: outer-loop verification systems, Just-In-Time (JIT) Context Hubs with semantic routing, cloud-native sandboxed code execution, and strict human-in-the-loop (HITL) guardrails. You build enterprise-grade agent platforms optimized for 2026 model economics (Luna, Sol, Muse Spark), bridging directly to Operating Systems, Distributed Systems, and Verification Compilers.
Week by week
Every week unlocks the next. Concepts route you to free, world-class material; projects turn that knowledge into something deployed.
Week 1
Engineer deterministic execution harnesses, multi-turn agent run-loops, and structured context state management separating core reasoning from environment orchestration.
Bridges to Operating Systems — processes, scheduling, and the run loopBuilds on: nothing, start here
Week 2
Implement deterministic function calling interfaces and tool schemas, contrasting pre-compiled tool pipelines with runtime tool generation for low-latency agent execution.
Bridges to Software Engineering — interface design and API contractsBuilds on: Foundations of Harness Engineering & Loop Crafting
Week 3
Standardize client-server tool integration, decoupled state exposure, and dynamic context injection using the open Model Context Protocol (MCP) specification.
Bridges to Computer Networks — protocols, client-server architecture, and RPCBuilds on: Tool Use, Pre-Compiled Toolmaking & Function Calling
Week 4
Scale reasoning via test-time compute allocation, Process Reward Models (PRMs), Monte Carlo Tree Search, and dynamic thinking token budgeting for System-2 deliberative tasks.
Bridges to Artificial Intelligence — search, planning, and state-space reasoningBuilds on: Tool Use, Pre-Compiled Toolmaking & Function Calling
Week 5
Design and build Context Hubs that solve API and schema deprecation. Master Just-In-Time (JIT) Context Delivery using semantic routers to prevent context bloat and control inference costs of premium model tiers.
Bridges to Operating Systems — memory hierarchy, paging, and cachingBuilds on: Planning, Test-Time Compute & Deliberate Reasoning Scaling
Week 6
Build secure executing runtimes for autonomous agents. Utilize Google's stateful Interactions API models with background task execution, provision managed Linux sandboxes, and orchestrate OAuth 2.1-compliant secure tool environments to thwart prompt-injection sandbox escapes.
Bridges to Operating Systems — virtualization, namespaces, and process isolationBuilds on: Context Hubs, JIT Context Delivery & Semantic Routers
Week 7
You are handed an enterprise mandate: the research division needs a launched product — a system that takes an open-ended technical question, autonomously researches it across many sources, synthesizes a defensible report, and ships the report as a published artifact, with zero human steps in the middle. Build an orchestrator-worker multi-agent system: a lead agent that decomposes the question and spawns specialized worker agents (search, read, synthesize, fact-check), coordinates their results through shared state, and produces a cited deliverable. This is not a notebook demo. The result must be a directly deployable, hyperscalable product: real public hosting, CI/CD on every commit, observability, security hardening, a polished and accessible web UI a non-technical analyst will happily use, and complete go-to-market material — a landing page, a pitch, and a recorded demo. The architecture must absorb concurrent research runs without falling over, and recover from a failed worker. We are not here to babysit the run; ship it as a real product.
Why it matters: Multi-agent research and synthesis systems are being deployed across consulting, finance, and R&D to compress weeks of analyst work into hours. Shipping a coordinated, fault-tolerant agent fleet makes a builder ready for an Agentic Systems Engineer or Applied AI Engineer role, where the bar is production reliability, not a demo.
A publicly hosted product with its own domain or stable URL, plus a public repo: the orchestrator and worker agents, an MCP-based tool layer, a fast accessible web UI for submitting questions and reading results, CI/CD running lint/tests/build on every commit, persisted and inspectable run traces, a marketing landing page, a 10-slide pitch, a recorded demo video, and a README documenting the coordination design, the failure-recovery and scaling strategy, and three example end-to-end runs with their published reports.
Market signal, who wants thisAgentic deep-research is one of the hottest 2026 categories: the AI agent market is projected to grow from $7.84B in 2025 to $52.62B by 2030 (41% CAGR), and a16z reports a portfolio pivot from copilots to autonomous systems, with Sierra, Glean, and Decagon as comparables and YC W26 funding multi-agent orchestration startups such as Tensol and Korso. Consulting, finance, and corporate R&D teams are actively buying systems that compress weeks of analyst work into hours; investors fund this because it sells time back to high-cost knowledge workers.
Week 8
Design multi-agent swarms using standardized Agent-to-Agent (A2A) protocols, decoupled task delegation, structured message-passing, and Manage-Execute-Audit verification state machines.
Bridges to Distributed Systems — coordination, message passing, and consensusBuilds on: Sandboxed Execution, Stateful APIs & Runtime Security
Week 9
Deep dive into programmatically compiling agent behaviors and declarative skills. Design self-evolving execution loops and testing harnesses that enable recursive self-improvement based on verification feedback.
Bridges to Compilers — program optimization, intermediate representations, and iterative refinement loopsBuilds on: Planning, Test-Time Compute & Deliberate Reasoning Scaling
Week 10
Architect strict Human-in-the-Loop (HITL) guardrails and automated verification outer loops. Replace fragile 'absolute autonomy' with deterministic testing harnesses that validate code-review safety and runtime constraints before commit.
Bridges to Software Engineering — testing, regression suites, and observabilityBuilds on: Multi-Agent Orchestration, A2A Protocols & Verification Loops
Week 11
An autonomous agent with tools is an attack surface. Prompt injection, tool poisoning, the lethal trifecta of private data plus untrusted content plus exfiltration, and least-privilege agent design.
Bridges to Information Security — threat modeling and least privilegeBuilds on: Agent Verification Loops & HITL Guardrails
Week 12
Design a highly resilient, cloud-native asynchronous execution harness. Build an inner loop that generates and modifies complex repositories inside stateful sandboxes, and an outer loop that validates compilation, runs tests, and applies strict Human-in-the-Loop (HITL) gates before committing changes.
A complete, production-grade asynchronous agent platform with remote container sandboxes, a semantic-routing JIT Context Hub, and an automated verification harness with a web-based approval interface.
What's next
Each track stands alone, so there's no wrong order. If you want a suggestion, this one pairs well next.