Leverage AI fully across your entire development lifecycle

Goal: Set a baseline level of knowledge

• Map AI usage maturity across teams.
• Identify foundational risks and opportunities in how AI is used.
• Awareness of foundational threats and principles (including shadow AI).
• Secure AI-assisted development
• Agents, orchestration including MCP, and safer AI integration into engineering workflows.
• Hands-on angles for the tools your team already uses (e.g. Claude Code, Cursor).

Goal: Equip the team and set up effective processes.

• Tooling battle: compare available tools by price/performance.
• Rules for secure usage in a team setup.
• Environment configuration: IDE setup aligned with company standards.
• Sub-agents: practical inspiration for specialized agents in different delivery phases.

Goal: Sustain and grow AI adoption across teams over the long term

• Identify people across teams who will act as ambassadors spreading best practices further.
• Internal ambassador roles, sharing know-how, and supporting adoption of standards.
• Unify best practices across the company.
• Targeted ambassador development

Goal: Accelerate delivery and analysis phases.

• Deep research & PRD: requirements, research, best practices, and handling client feedback.
• Legacy code analysis: understand unfamiliar projects in minutes.
• Case study: finding 120+ app speed improvements in 2 hours.
• Security & QA: AI for security audits and first-layer code review.

Goal: Systematically measure and improve the quality of AI outputs in your workflows.

• Metrics and checklists for scoring model responses.
• Benchmarks and comparisons of prompt or model versions.
• LLM as a judge: a secondary model as an independent scorer (scores, rubrics, brief feedback on responses).
• Combining automated checks with human validation.
• Practices for catching hallucinations and regressions after changes.

Goal: Connect internal tools and data to AI workflows via the Model Context Protocol.

• MCP fundamentals and common server patterns.
• Secure access to internal APIs and data sources.
• Local development, testing, and deployment of MCP servers.
• Integration with IDEs and agents your team already uses.

Goal: Search company documents and codebases with strong retrieval accuracy.

• Chunking, embeddings, and choosing a vector database.
• Designing ingestion pipelines and refreshing the knowledge base.
• Trade-offs between latency, cost, and answer quality.
• Re-ranking, source citations, and foundational guardrails.