Working with AI coding tools
Use Codex, Claude Code, Cursor, and GitHub Copilot on development tasks. Learn how to give an agent useful context and break a task into work you can check.
Corporate AI Training
Practical AI training. Built on real work.
I teach teams how to use Codex, Claude Code, Cursor, and GitHub Copilot, drawing on the same tools I use to build and maintain production AI applications.
Previous sessions
Teams I’ve trained.
Corporate sessions covering the practical use of AI tools for software development.
Mayank Bohra · AI engineer at Gradeless AI · AI Manager Academy
Workshop topics
Start with your team’s work.
From AI coding fundamentals to advanced agent workflows and team governance, shaped around your team’s tools and experience.
Debugging, testing, and review
Work through a bug, ask an AI tool to propose a change, and assess the result. Use tests and code review to decide whether a change is ready to ship.
Custom skills and project instructions
Turn recurring tasks into reusable skills. Write project instructions that capture your team’s conventions, add examples and reference material, and test whether an agent follows them on real tasks.
Plugins, hooks, and shared tooling
Build and configure plugins that bring skills, agents, and integrations together. Use hooks for repeatable checks, review what a plugin can access or execute, and manage updates to shared team tooling.
MCP and connected workflows
Connect agents to documentation, developer tools, and approved sources of context through MCP. Work through server setup, authentication, scoped access, and troubleshooting failed tool calls.
Agent workflows and orchestration
Design workflows that move from planning to implementation and review. Delegate focused tasks to sub-agents, manage context and handoffs, and use checkpoints to assess the work before it moves forward.
AI governance and team adoption
Define which tools and data an agent can use, which actions need human approval, and who owns the resulting changes. Cover permissions, sensitive data, plugin and MCP review, audit trails, and practical rules for adopting AI across a team.
From production to practice
Engineering informs the teaching.
At Gradeless AI, I build Rehearsal AI’s voice interview systems. My projects include Highlyt and MCP developer tools, and my merged open-source contributions include Resend and Deepgram SDKs. These give me practical examples of debugging, testing, and reviewing AI-assisted work.
Explore my engineering workOpen-source contributions
Merged changes to Resend and Deepgram SDKs. Each link includes the code, discussion, and merge history.
AI Manager Academy
I’m affiliated with AI Manager Academy (AMA), where I also write about AI tools, building applications, and changes in how we work.
Read AI Manager AcademyPlan a session
What should your team practise?
Share your audience, the tools they use, what you want them to practise, and your preferred dates. We’ll agree on the topic, scope, and format.
Email Mayank