Crash course: Agentic Engineering for "Greenfield" projects
Mentor - Vyacheslav Koldovskyy, Competence Manager at SoftServe
Why do some AI-first projects take off, while others remain demos? The secret is not only in the tools, but in how the engineering around them is built.
Claude Code and OpenClaw are vivid examples of how the greenfield approach allows you to use the potential of agents to the maximum: lay the right architecture, think through the rules of interaction, connect the necessary tools and move to production at a completely different speed.
In this course, we will analyze the approaches behind such AI-first solutions and learn how to apply them in practice to create your own production-ready products.
Mentor - Vyacheslav Koldovskyy, Competence Manager at SoftServe
Important: How to work with AI agents correctly?
- ❌Wrong:
- chaotic use of agents in "vibe coding mode" quick start without a well-thought-out architecture and optimal technical stack
- code generation without specifications and acceptance criteria
- parallel work of agents without coordination
- accelerated delivery at the cost of technical debt
- lack of reliable guardrails and review mechanism
- ✅ What works:
- conscious choice of technologies, frameworks and architecture for AI-first delivery
- building a system of rules, roles and skills for agents from the very beginning
- finding a balance in connecting the necessary MCPs for external tools, environments and data sources
- Spec Driven Development as the basis of managed implementation
- multi-agent orchestration for parallel work on different parts of the system
- a seamless path from idea and specification to production
- new development processes for a new era of engineering with AI agents
With the right approach, AI in greenfield projects provides not just “faster code generation”, but a fundamentally different pace of product creation: when the team focuses on functions, business logic and solutions, and a significant part of the technical implementation is taken on by agents. You will learn to manage your agents locally, in the cloud, and even from a smartphone and will be able to deliver features from anywhere and anytime.
The course will be practical, we will consistently build a full-stack web project from scratch, using the best modern practices of Agentic Engineering and even embed an agent based on OpenClaw / NanoClaw into our solution.
During practical tasks, you will have the opportunity to implement your own idea or startup as a finished product.
For the demo we will use Cursor, but in general the approaches are universal and suitable for any Agentic IDE, in particular Claude Code / GitHub Copilot, etc.
- start a new AI-first project and deploy it to production on the first day
- scale up the product through Spec Driven Development with rules, skills, MCP and multi-agent orchestration
- maintain quality and avoid technical debt through guardrails, code review and fast feedback loops
- integrate an agent based on OpenClaw / NanoClaw into your product and translate individual features into the Agent Skills format
- accelerate delivery to production without losing quality and manageability
After the course, you will be able to:
Format:
Crash course: 3 online sessions
Dates:
June 23 and 25: 18:30-21:00
June 27: 11:00-14:00
Duration of each session: ~3 hours.
Platform: Zoom. You will receive a link to the broadcast before the course starts, and it will also be available on this landing page. In addition, all materials and lecture recordings will be available on the learning platform.
Preparation: Before the start of the course, participants will receive instructions for preparing the environment, tools, and a basic set of services for work.
Practical task: To consolidate the material and receive a certificate, participants must create a project in accordance with the requirements of the course program, which will be checked by a mentor (deadline: one week after the last session).
Event language: Ukrainian
Presentation language: Ukrainian with English terms
Who will be interested?
- Developers of all levels who want to launch new products with AI not chaotically, but through an engineering approach that meets the best modern practices.
- Tech Leads, Solution Architects and Founding Engineers, who design new systems and want to immediately adapt them for effective agent work.
- Startup founders and product builders who want to shorten the path from idea to working product and production deployment.
- DevOps / Platform Engineers, who want to build a fast delivery pipeline for AI-assisted and agent-driven development.
- Engineering Managers, mentors and team leaders, who want to standardize work with AI tools on new projects.
- Product managers and business analysts who want to independently implement their ideas and release them into production.
Program
- The mentor leads one demonstration project during the course, and each participant simultaneously develops their own product using the same approach.
Day 1: Ship the first working product
- We start our end-to-end Next.js project, define an AI-friendly structure and bring it to the first deployment on Vercel by the end of the class.
- Why greenfield is the best scenario for Agentic Engineering and how to make "ship on day one" a reality
- How to move from an idea to a short heat and an AI-friendly repository structure
- Next.js + Vercel as the default for AI-first greenfield: when exactly this stack, and when another
- Basic engineering context, rules and connecting Vercel React best practices skills
- How to develop mobile-first and use AI agents as multipliers outside the desktop
Day 2: Growing the product through SDD, backend, and guardrails
- Adding backend, features, and discipline. We practice Spec Driven Development, expand the rules/skills/MCP system to project tasks and stitch the process with code review and other guardrails.
- Spec Driven Development as the basis of managed development and the Plan → Spec → Implement → Verify cycle
- Roles for agents, custom Agent Skills for your domain, MCP for external tools and data sources
- Multi-agent orchestration: task distribution between agents and compilation of results into a single system
- Adding a backend to a Next.js project under the control of agents
- Guardrails: automatic code review, quality control, protection against technical debt; fast feedback loops and deployment in production
Day 3: Embedding an agent based on OpenClaw / NanoClaw
- We show the skill-based programming paradigm in action. We translate one of the features implemented on Day 2 with regular deterministic code into an agent based on OpenClaw / NanoClaw and integrate it with our web interface - we get a hybrid architecture "web + agent-runtime".
- OpenClaw / NanoClaw / NemoClaw: a brief overview of the approaches and when to choose which
- Skill-based programming as a paradigm: when business logic lives better as Agent Skills, not as code
- From a deterministic feature to an Agent Skill: rewriting a piece of code from Day 2 into a skill format
- Integrating the agent with the web interface; optionally — additional channel via messenger
- Security, models, cost and limits: what to look for when placing an agent in the market
Mentor: Viacheslav Koldovskyi
— Competence Manager at SoftServe
— 20+ years in IT, certified Google Cloud Professional Cloud Architect, nVidia Generative AI LLMs
— Ph.D / Candidate of Economic Sciences, Associate Professor, Head of the Gen AI Center at IT STEP University
— Founder of YouTube and Telegram channels Programming Mentor
— Active speaker: speaking at iForum, DOU Day, etc.
— Leader of the AI community at DOU, author of publications
— Has implemented projects with Gen AI and AI-generated code, which successfully working in production
— Advises companies on transforming SDLC processes using AI
Event price
Access to 3 online broadcasts (~10 hours)
Presentations and materials
Access to the course recording on the learning platform (1 year)
Access to the chat in the TG, where the speaker will answer questions (1 month)
Participant certificate (subject to completing homework)
Free month in Fwdays Club
Course participants receive a 10% discount on participation in Fwdays conferences
Affordable payment in installments from Monobank and purchase in installments from PrivatbankIf after the first lesson you realize that your expectations do not match, we will refund the money in full.