Meera Pillai

Author

AI developer at a Trivandrum health-tech company, building diagnostic support tools using multimodal LLMs. Integrated the FDA-cleared ML pipeline that assists radiologists in detecting lung nodules.

aillmapisecurity

Articles by Meera Pillai(2)

Detecting and mitigating prompt injection attacks

Prompt injection is the primary attack surface for LLM-powered applications — attackers embed instructions in user-supplied or retrieved content that override your system prompt, exfiltrate data, or cause the model to perform unintended actions, and this guide covers detection and defence strategies with working TypeScript examples.

securityprompt-injectionllmattacksmitigation

17 July 2026

How LLM tool use works: function calling under the hood

Tool use lets language models interact with the world by invoking functions defined by the developer. This article tears apart the request-response cycle, JSON schema negotiation, parallel tool calls, and the pitfalls every production integration hits.

llmfunction-callingtool-useapiai-engineering

16 July 2026