Welcome to Robert’s Bytes

Scientific writing largely borrowed from my dissertation. I graduated with a PhD in cardiac systems biology from the Saucerman lab and my research revolved around understanding how genetics impacts heart failure and how drugs are effective in different patients. My day job is working in data and ML at a fintech.

Integration of human transcriptome and cardiac systems models

Study of the impact of signaling pathway alterations in heart disease

June 30, 2023 · 14 min · Robert Amanfu

ReconBench: benchmarking LLMs on cardiac signaling network reconstruction

An Inspect-native replication and extension of Tewari et al. (2025). Recall reproduces, precision and F1 are new, and the main failure traces to an extraction format mismatch rather than a reasoning gap.

April 27, 2026 · 8 min · Robert Amanfu

Beyond sequence similarity: function-aware screening of synthesized DNA

Protein and DNA foundation models add a function-level screening layer that is robust to disguise. An ensemble of ESM-2 and Evo2 7B catches 85.8% of toxins at a 1-in-100 false-alarm rate, complementing, not replacing, the alignment-based policy screens used today.

April 24, 2026 · 8 min · Robert Amanfu

Treasury Reapers: Sentient Arena Challenge 0 final report

Final report for my Treasury Reapers team submission to Sentient Arena Challenge 0: a goose-based OfficeQA agent, prompt and skill-contract iterations, leaderboard progression, and lessons from trace-driven optimization.

April 8, 2026 · 7 min · Robert Amanfu

Porting scBench to inspect_evals: lessons from running AI agents on single-cell data

Lessons from moving scBench’s public task set into the inspect_evals framework, including scoring rules, evaluation setup, and what run logs revealed.

April 2, 2026 · 13 min · Robert Amanfu