The Hidden Fragility of AI Safety
How guardrails can obstruct defense and weaken system resilience.
Software Research Engineer
Dionizije Fa
I wasn’t scared to look at the data, so I ended up in the eval mines.
Latest Research Topics
Beyond Benchmark Scores: An Analysis of Bioinformatics Agent Evaluations
Tradeoffs in bioinformatics evals and how to mitigate them.
Harness Bench
Benchmarking popular agent harnesses.
Sound agentic science requires adversarial experiments
The only real verifiers are experiments.
BioAgent Bench
Evaluation suite with adversarial testing for end-to-end bioinformatics tasks.