← Back to Projects
Python
AI Evaluation
Agent Reliability
Adversarial Testing
Python
AI Evaluation
Agent Reliability
Adversarial Testing
Project
Autonomous Agent Reliability & Assurance Platform
M↓ABOUT.md
M↓EXPERIENCE.md
M↓PROJECTS.md×
1
# Projects
2
3
Selected engineering and machine-learning projects.
4
5
## Autonomous Agent Reliability & Assurance Platform
6
Adversarial evaluation framework scoring autonomous agent reliability across reasoning, tool use, and recovery.
8
9
## GPT-2 Fine-Tuning
10
Domain-adapted a pretrained GPT-2 model using PyTorch and Hugging Face Transformers.
12
13
## LLM Agent with Tool Use, Memory, and Retrieval
14
A ReAct-based agent with persistent memory and retrieval-augmented generation, built on DSPy.
Tools & technologies
Python
AI Evaluation
Agent Reliability
Adversarial Testing
Python
AI Evaluation
Agent Reliability
Adversarial Testing
Stress-testing autonomous agents before they ship.
An adversarial evaluation framework that scores agent reliability across instruction adherence, tool selection, and multi-step reasoning — surfacing the failures that only show up under pressure.
Adversarial
Test environments
Automated
Scoring pipelines
Human-in-the-loop
Oversight flags
Built a rigorous evaluation framework for autonomous AI agents, measuring instruction adherence, tool-selection accuracy, multi-step reasoning, failure recovery, and factual verification. Designed adversarial test environments and automated scoring pipelines to expose subtle reliability failures, benchmark agent architectures, and identify where human oversight is required.
More Projects
Explore more of my work.
$ python train.py --model gpt2 --epochs 3
Epoch 1/3 [###.......] loss: 2.184
Epoch 2/3 [#######...] loss: 1.607
Epoch 3/3 [##########] loss: 1.203
GPT-2 Fine-Tuning
class Agent:
def step(self, query):
thought = self.reason(query)
return self.memory.recall(action)