AI-Assisted Engineering
InfoQ Launches AI-Assisted Engineering Certification for Senior Developers
A new five-week online program teaches experienced engineers how to safely deploy and verify coding agents in production environments.
InfoQ
InfoQ is now accepting participants for its AI-Assisted Engineering certification program, taught by seasoned practitioners working with agentic AI systems at scale. The inaugural cohort begins September 18 and runs through October 16, 2026, with sessions on Friday mornings; a second group starts in October. The course costs $1,470 and targets senior engineers and architects who already deploy AI coding agents to real production systems.
Who Should Enroll#
The program is designed for experienced developers working with established codebases—not toy projects or proof-of-concept experiments. Participants will learn what coding agents can and cannot safely do, how to spot when they fail, and how to demonstrate to stakeholders that AI-generated code meets organizational standards for quality, performance, and maintainability.
Zichuan Xiong, Head of AIOps at Thoughtworks and creator of agentic systems for software operations, observes that "Teams are adopting coding agents faster than they're building the verification around them." The course addresses this gap by building what Xiong calls a "harness"—the infrastructure surrounding the model, including access controls, monitoring, approval workflows, and automated checks that enable an agent to self-correct before a human reviews the change.
Weekly Curriculum#
Over five weeks, participants attend four-hour live sessions each week, working through frameworks applied to a shared, real-world codebase alongside senior engineers from other organizations. Each week begins with a pre-recorded QCon talk providing conceptual foundation.
- Week 1: Bringing an agent into an unfamiliar project with minimal permissions and isolated environments.
- Week 2: Converting vague issue descriptions into verifiable specifications for both agents and reviewers.
- Week 3: Implementing independent review mechanisms so the generating agent cannot evaluate its own output.
- Week 4: Embedding verification checks into continuous integration pipelines to support unattended runs.
- Week 5: Turning observed review patterns into reusable rules; capstone presentations comparing actual results to initial predictions.
