AI applications are increasingly moving beyond simple LLM calls. Modern AI systems often need to make thousands of small decisions — Which agent should handle this request? Should this tool call be allowed? Is this document relevant? Which action should an agent take next? Should this content be approved or rejected? This is where Jev introduces an interesting new approach. In this hands-on meetup, we'll break down what Jev is, how it works, where it fits alongside traditional LLMs, and how developers can use it in real-world AI systems. Rather than treating Jev as another replacement for an LLM, we'll understand it as a decision layer that can work alongside your existing AI stack.
What You'll Learn
- What exactly is Jev and why was it introduced?
- How Jev is different from traditional LLMs
- Understanding Jev's decision primitives
- How Jev handles Choice, Score, and Yes/No decisions
- Understanding probabilities and confidence
- When to use Jev vs an LLM vs traditional ML/classifiers
- How Jev fits into AI Agent architectures
- Using Jev for agent routing and decision-making
- Using Jev for guardrails and AI verification
- Jev for classification, triage and structured decisions
- Jev in RAG and semantic search workflows
- Using Jev in real-time decision loops
- Designing production-ready Jev workflows
- Understanding where Jev should NOT be used
Production Use Cases
We'll explore practical scenarios such as:
- 🤖 AI Agent Routing — deciding which agent should handle a task
- 🛡️ AI Guardrails — verifying inputs, outputs and tool calls
- 🔎 RAG & Search — deciding which information is relevant
- 📊 Classification at Scale — processing large volumes of structured data
- 🎯 Task & Ticket Routing — automatically categorizing incoming requests
- 🌐 Browser / Computer Agents — deciding the next action
- ⚡ Real-time Decision Systems — making fast decisions inside an application loop
- 🧪 AI Evaluation — judging AI-generated outputs against defined criteria
These patterns are already appearing in the early Jev ecosystem, although many publicly shared examples are still prototypes or early-stage implementations rather than established production case studies.
Who Should Attend?
This meetup is ideal for:
- Software Engineers
- Backend & Frontend Developers
- AI Engineers
- Developers building AI Agents
- Python Developers
- Developers exploring LLM applications
- AI enthusiasts with programming experience
Basic programming knowledge is recommended. No prior Jev experience is required.
Format
- 🧑💻 Live Coding & Demo session
- 💬 Interactive Q&A
- 🏗️ Real-world architecture discussions
- 🚀 Production use-case walkthroughs
The focus is on understanding and building, not just watching. We'll start with “WTF is Jev?” and end with a practical understanding of where this new decision layer can fit into production AI systems.
Duration: 1.5 Hours
