Tisha Chawla

Tisha Chawla is a Software Engineer at Microsoft, where she builds agentic systems engineered to run reliably against production data. Her work centers on the structural foundations of agent infrastructure: durable state management, deterministic execution frameworks, and self-healing orchestration layers that scale beyond isolated demos. Alongside Susheem Koul, she is the co-creator of AgentPlane, an open-source infrastructure layer built for teams deploying production-grade agents. Tisha is a published researcher with peer-reviewed papers in applied machine learning across venues like APNET (SIGCOMM) and ASONAM. A frequent speaker on developer productivity and deterministic AI architectures, she recently presented at the AI Engineer World's Fair 2026 in San Francisco and regularly delivers deep-dive technical sessions to engineering teams across Microsoft.

Microsoft, Indiatishachawla.netlify.app tishachawla-jg linkedin.com/in/tisha-chawla

Interests:

machine learning productivity artificial intelligence

Debugging the Agent That Refuses to Fail the Same Way Twice

AI Coding Summit NYC

Upcoming

Debugging the Agent That Refuses to Fail the Same Way Twice

Your coding agent botched a multi file refactor once in production. You rerun the exact same prompt to debug it, and it works. The model sampled different tokens, a tool returned a different payload, the context drifted, and the failure was gone. This session shows how to capture every nondeterministic input to an agent loop into an append only execution log, replay the failed run step by step, and then turn that replay into reproducible evals so the bug never quietly returns.

artificial intelligence