NAST

Nazmus Sakib Touhid

Lecturer

nazmus.touhid@bracu.ac.bd

Address

CSE Department
4th floor, Room No # 4P195,
Brac University,
Kha 224 Bir Uttam Rafiqul Islam Avenue,
Merul Badda, Dhaka, Bangladesh

  • Passing Without Repairing: Verifier Blind Spots in a Live SRE Agent Benchmark With the Service deleted and no agent acting, the benchmark's grader still reports the fault as mitigated, because it only checks Deployment and Pod health. Accepted at NeurIPS 2026 Workshop "Who Verifies the Agents?"
  • MemAgent: A cache-inspired framework for augmenting conversational Web Agents with task-specific information. Published in WOA 2025 – 26th Workshop "From Objects to Agents"
  • Luck Is Not Skill: When Do Paired Rollouts Help Group-Relative RL of LLM Agents? Giving every rollout in a GRPO group the same luck lifted success under tool faults from 61.8% to 66.8%, but not under grader noise. Arxiv-Preprint
  •  
  • UIUC++ Summer Research in Software Engineering (Summer 2026, Research Intern)
  • SWE, priyo.com (2025)
  • AI for SWE & SRE 
  • LLM post-training
  • Reinforcement Learning
  • AI safety
  • Agent orchestration

Thesis

Accepting


As:

Supervisor

Level:

Undergraduate

Type:

  • Thesis

Research Interest

  • Natural Language processing
  • Reinforcement learning
  • Machine Learning
  • Deep learning

©2026 BracU CSE Department