I work on AI Agents and Model evaluations, with a particular focus on model behaviour, agentic decision-making, adversarial robustness, and red-teaming.
Looking for AI researcher and research engineering roles. Open to relocation or remote.
Evaluates how open-weight language models respond to harmful requests written in non-canonical forms such as emojis, leetspeak, homoglyphs, Base64, and invisible Unicode. Introduces the Adversarial Surface-Form Robustness Dataset (ASRD) and the Quad-State Evaluation Rubric (harmful compliance, safe response, comprehension failure, indeterminate).
NeurIPS 2026 Workshop (EvoRobust)
Lookup Is Not All You Need
arXiv soon
Pavan Maddula
Evaluates whether tool-using AI agents check the available information before committing to an action in high-stakes environments across military, cybersecurity, healthcare, and governance.
Safety evaluations of LLMs predominantly use text-based adversarial prompts. This paper evaluates 50 emoji-augmented prompts on four open-source models as a test of that gap.
M P V S Gopinadh, Karthik Kamuju, Kummari Avinash, Muppana John Joshua, Srinivasa Raju Rudraraju
In a controlled synthetic setting, standard RLHF aggregates heterogeneous preferences into a single reward model. When preferences conflict, majority modes can dominate reward learning, leaving minority preferences under-represented.
Built an AI-powered recruitment system using agents and automations for candidate sourcing, resume screening, and evaluation, cutting manual recruitment effort and turnaround time.
Developed a vendor research agent that aggregates client information, tracks news, and surfaces structured insights - conferences, keynotes, product launches, strategic updates.
Designed the action plan and architecture for an in-house talent pool platform using LLM + RAG pipelines for long-term candidate retention and intelligent discovery.
Collaborated with cross-functional teams to integrate these systems into existing workflows and translate business needs into agent-based solutions.
Internal Smart India Hackathon Winner 2025. Application-layer email encryption using IBM quantum key material, post-quantum crypto, and AES over standard SMTP.
Amaravati Quantum Valley Hackathon Winner 2025. Hybrid quantum–classical router that turns perishable-goods delivery into a QUBO and solves it with QAOA.