The AI Forge · Advanced · F3
Agentic Systems & AI Apps
Agent design, tool use, MCP, multi-agent orchestration, prototype to production
- Duration
- 3 days · 9:00–17:00 each day (19.5h contact, incl. breaks + lunch)
- Participant level
- Advanced: experienced practitioners
- Format
- Instructor-led labs, build-along
- Participants
- 8 included · 12 maximum (flexible group size)
- Prerequisite
- F1 + F2 (or equivalent)
- Ecosystem
- Agnostic: any model provider or stack
- Price
- EUR 11,500 (EUR 650 per extra participant)
Amounts in EUR, excl. VAT.
ai-generatedWhat participants will be able to do
- Design multi-agent systems with tool use, memory, and planning
- Build MCP-based agentic workflows in production
- Implement agent safety, guardrails, and human-in-the-loop controls
- Deploy and monitor production AI agents end to end
Tools & resources
What changes after this module
Engineers design, build and productionise agentic applications, tools, memory, multi-agent orchestration and a standard tool-connection protocol, grounded in patterns that survive framework churn.
Who should attend
Engineers and data professionals. Advanced: experienced practitioners.
Programme
This agenda is indicative. Content, sequencing, and examples are adapted to your team's context, tools, and objectives.
Before you start
- A laptop with a working development environment and admin rights
- Proficiency in at least one general-purpose language (Python and/or JavaScript/TypeScript)
- A code editor/IDE with an approved AI coding assistant enabled
- An API key for an approved model provider (issued by the company)
- A Git repository and command-line comfort
- Completion of F1 + F2 or equivalent
- A real task with tool/API access to turn into an agent
| Day 1 | 9:00–10:30 | Agent = model + harness: What separates an agent from a chatbot; the autonomy spectrum; why the harness (tools, state, control loop) is where reliability lives Lab: Classify candidate tasks on the autonomy spectrum and sketch the harness for the one real task you’ll build across the three days. |
| 10:30–10:45 · ☕ Coffee break | ||
| Day 1 | 10:45–12:30 | Tool use & function calling: Defining tools, argument validation, error handling and keeping deterministic work out of the model Lab: Give your agent 3 real tools with validated inputs, route deterministic steps to plain code, and treat every tool response as untrusted input (the tool-poisoning line). |
| 12:30–13:30 · 🍽 Lunch break | ||
| Day 1 | 13:30–15:30 | A standard tool-connection protocol (MCP): Connecting agents to tools/data through an open protocol instead of bespoke glue: a durable interface Lab: Connect your agent to a tool/data source via an MCP-style server and call it end-to-end. |
| 15:30–15:45 · ☕ Coffee break | ||
| Day 1 | 15:45–17:00 | Lab: single agent: Assemble a working single agent with tools Lab: Ship a single agent that completes a real multi-step task using its tools. |
| Day 2 | 9:00–10:30 | State & memory: Short- and long-term memory, context compaction and why agents fail as state-management failures Lab: Add persistent memory + context compaction and show the agent handling a long task without losing state. |
| 10:30–10:45 · ☕ Coffee break | ||
| Day 2 | 10:45–12:30 | Multi-agent orchestration: Supervisor/worker, hand-offs and when NOT to go multi-agent (monolith fragility vs coordination cost) Lab: Decompose a task into a supervised multi-agent workflow and compare it against the single-agent version. |
| 12:30–13:30 · 🍽 Lunch break | ||
| Day 2 | 13:30–15:30 | Evaluating agents: Trajectory and outcome evaluation, tool-call correctness and regression testing for non-deterministic flows Lab: Build an agent eval (trajectory + outcome) on your task’s success set: defined at the end of Day 1: and catch a regression you introduce on purpose. |
| 15:30–15:45 · ☕ Coffee break | ||
| Day 2 | 15:45–17:00 | Lab: orchestrated system: Wire the multi-agent flow with evals Lab: Ship the orchestrated system and run it against your eval set. |
| Day 3 | 9:00–10:30 | From prototype to application: Turning an agent into a real app: APIs, UI/surface, human-in-the-loop checkpoints, guardrails Lab: Wrap your agent in an API + minimal UI with a human-approval checkpoint on risky actions. |
| 10:30–10:45 · ☕ Coffee break | ||
| Day 3 | 10:45–12:30 | Reliability & observability: Tracing tool calls and decisions, timeouts, retries, fallbacks and cost control for agents Lab: Add tracing + fallbacks and use a trace to debug a deliberately broken tool call. |
| 12:30–13:30 · 🍽 Lunch break | ||
| Day 3 | 13:30–15:30 | Lab: production agentic app: End-to-end deploy of the agentic application Lab: Deploy the agentic app with tracing, guardrails and human-in-the-loop live. |
| 15:30–15:45 · ☕ Coffee break | ||
| Day 3 | 15:45–17:00 | Review & hardening: Peer review, failure-mode walkthrough, next steps Lab: Present the app, walk its failure modes, and agree the hardening backlog. |
Deliverables
- Deployed agentic application (API + UI + human-in-the-loop)
- Tool integrations via a standard protocol
- Multi-agent orchestration with memory
- Agent eval suite (trajectory + outcome)
- Tracing + guardrails + hardening backlog
Interested in running this module for your team? Get in touch and we'll tailor the format, dates, and delivery to your context.
Request this trainingYour trainer
Senior expert
Builds agentic systems for production; hands-on MCP and multi-agent orchestration
Discovery Session
This module requires a full day to deliver a meaningful learning experience. We only offer it in its complete format.
Frequent questions
Can modules be taken individually?
Yes. Every module stands alone at a fixed price, and every module counts toward a programme if you continue.
Where does training happen?
At your premises or remote, on your dates, for private cohorts. Open sessions run on a fixed monthly calendar.
Which AI tools do you train on?
Yours. Every module ships in four ecosystem editions and runs its exercises on your real stack.
Who delivers?
Inforca's senior consultants and trainers. Flagged modules and the executive track are delivered at senior-expert level.
Discuss this module in a First Call: fit, dates, and the path around it.
No commitment · our team responds within one business day