PromptOps
Today
Practice
Library
Exam
Certificate
Search
⌘K
ខ្មែរ
◐
Sign in
Today
Practice
Library
Profile
Today
Track 6
6. Autonomous Agents & System Design
0 / 22 lessons
Start track
1
Agentic Prompting & Tool Use
Prompts that command autonomous pipelines
Advanced
2
Prompt Testing & Version Control
Treat prompts like production code
Advanced
3
AI-Assisted Decision Making
Structure your judgment, don't replace it
Advanced
4
Prompting Coding Agents
Write a job description, not a wish
Intermediate
5
Computer-Use & GUI Agent Prompting
It has your mouse — write the guardrails first
Advanced
6
Prompt Compilers & DSPy
Stop hand-tuning. Start compiling.
Expert
7
Constrained Decoding & Structured Outputs
Guaranteed structure. Zero parsing errors.
Expert
8
Multi-Agent Orchestration & Protocol Prompting
Prompting the team, not the individual.
Expert
9
Multi-Agent Systems & Handoffs
Divide cognition. Coordinate action.
Expert
10
Prompt Drift & Calibration
Detect and correct changes in model behavior over time.
Expert
11
Prompt Red-Teaming & Vulnerabilities
Audit prompt injection vulnerabilities in system instructions.
Expert
12
Prompt Optimization Algorithms
Automate prompt refinement using LLMs as optimizers.
Expert
13
Agentic Workflows
Design prompt-driven loops for autonomous multi-agent systems.
Expert
14
Prompt Versioning & Git Workflows
Manage prompt templates as code artifacts in version control pipelines.
Expert
15
Programmatic Pipelines (DSPy)
Build dynamic, code-driven prompt structures rather than static text files.
Expert
16
Prompt Ensembling via MoE
Route prompt segments to specialized models and merge outputs.
Expert
17
Prefix-Tuning & Soft Prompts
Optimize continuous virtual token embeddings instead of hard text prompts.
Expert
18
Self-Correction in Compilers
Feed compiler errors back to the model to automate syntax fixing.
Expert
19
Memory-Guided Agent Prompts
Maintain structured long-term memory logs to guide multi-turn agent tasks.
Expert
20
Structured Output Contracts
Enforce strict typing and validation constraints on model responses.
Expert
21
Meta-Prompting (APE)
Use models to generate, critique, and optimize prompt templates.
Expert
22
Semantic Routing Pipelines
Direct user queries to specialized prompts based on embedding similarity.
Expert
6. Autonomous Agents & System Design