Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m a B.Tech CSE student focused on building secure, production-grade ML/LLM systems and agentic AI pipelines. I’ve designed and scaled ML security workflows to detect adversarial behaviors like model stealing, data poisoning, and evasion—improving detection coverage by 12%. I also enjoy shipping full-stack and AI-powered products: from low-latency C++ trading engines and TCP socket exchange simulations to React/TypeScript conversational platforms and RAG systems with injection/poisoning safeguards and automated faithfulness evaluation. Also I can create Machine Learning models and Do any task which uses AI and LLM.…I’m a B.Tech CSE student focused on building secure, production-grade ML/LLM systems and agentic AI pipelines. I’ve designed and scaled ML security workflows to detect adversarial behaviors like model stealing, data poisoning, and evasion—improving detection coverage by 12%. I also enjoy shipping full-stack and AI-powered products: from low-latency C++ trading engines and TCP socket exchange simulations to React/TypeScript conversational platforms and RAG systems with injection/poisoning safeguards and automated faithfulness evaluation. Also I can create Machine Learning models and Do any task which uses AI and LLM.

Ayush Parwal

AI Engineer, Data Scientist, AI Developer, +1





I’m a B.Tech CSE student focused on building secure, production-grade ML/LLM systems and agentic AI pipelines. I’ve designed and scaled ML security workflows to detect adversarial behaviors like model stealing, data poisoning, and evasion—improving detection coverage by 12%. I also enjoy shipping full-stack and AI-powered products: from low-latency C++ trading engines and TCP socket exchange simulations to React/TypeScript conversational platforms and RAG systems with injection/poisoning safeguards and automated faithfulness evaluation. Also I can create Machine Learning models and Do any task which uses AI and LLM.…I’m a B.Tech CSE student focused on building secure, production-grade ML/LLM systems and agentic AI pipelines. I’ve designed and scaled ML security workflows to detect adversarial behaviors like model stealing, data poisoning, and evasion—improving detection coverage by 12%. I also enjoy shipping full-stack and AI-powered products: from low-latency C++ trading engines and TCP socket exchange simulations to React/TypeScript conversational platforms and RAG systems with injection/poisoning safeguards and automated faithfulness evaluation. Also I can create Machine Learning models and Do any task which uses AI and LLM.

Available to hire

I’m a B.Tech CSE student focused on building secure, production-grade ML/LLM systems and agentic AI pipelines. I’ve designed and scaled ML security workflows to detect adversarial behaviors like model stealing, data poisoning, and evasion—improving detection coverage by 12%.

I also enjoy shipping full-stack and AI-powered products: from low-latency C++ trading engines and TCP socket exchange simulations to React/TypeScript conversational platforms and RAG systems with injection/poisoning safeguards and automated faithfulness evaluation.

Also I can create Machine Learning models and Do any task which uses AI and LLM.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Work Experience

ML Engineering Intern at Grafyn AI

July 1, 2025 - December 1, 2025

Designed and scaled ML security pipelines on AWS, Databricks, and Snowflake to simulate and detect model stealing, perturbation, and data poisoning attacks, improving adversarial threat detection coverage by 12%. Collaborated with Red Teaming to harden production agents against jailbreaking, prompt injection, obfuscation, and LLM hijacking using automated testing workflows. Built scalable data ingestion and evaluation pipelines using Python and Spark to monitor adversarial behavior and security metrics across large-scale ML workloads.