About us
"A team of androids without the mechanical arm."
I build virtual white-collar teams -- autonomous inference systems that do the work of analysts, researchers, and coordinators, then hand humans the results with receipts.
Founder of Aigentic -- institutional-grade AI research infrastructure where machines accelerate discovery and humans make decisions.
Aigentic ships human-supervised AI systems that replace headcount-heavy knowledge work with orchestrated inference pipelines. Not chatbots. Not copilots. Full workflow execution that moves numbers on a P&L.
| Layer | What it does |
|---|---|
| Inference Circuits | Modular agentic pipelines -- each "team member" is a purpose-built reasoning chain with tool access, memory, and accountability |
| Lorentzian Memory | Causal memory architecture grounded in relativistic geometry -- agents retrieve only what is semantically reachable, not everything that is similar |
| Looped Reasoning | Iterative transformer architectures that trade parameters for compute loops, getting more out of smaller models |
| Human-in-the-Loop Oversight | Every output is auditable. AI proposes, domain experts dispose. No black-box decisions reaching production |
Most AI companies sell acceleration. I sell replacement of entire workflows -- with measurable ROI, not vibes.
- Virtual analyst teams that run 24/7 research pipelines across financial, legal, and scientific domains
- Deliberately guided inference -- not "prompt and pray," but structured reasoning with verification gates
- Quantifiable gains to the bottom line: fewer FTEs on repetitive knowledge work, faster time-to-insight, auditable outputs
- Lorentzian memory lightcones -- replacing flat vector similarity with causal structure: memories live in a semantic spacetime where retrieval respects precedence, not just proximity. Polymathic weighting scores cross-domain relevance so agents surface unexpected connections instead of echo chambers
- Agentic orchestration -- multi-agent systems with tool use, persistent memory, and self-correction loops
- Looped language models -- challenging scaling laws by injecting iterative reasoning into inference rather than adding parameters
- Knowledge manipulation -- moving beyond retrieval (memorization) toward genuine reasoning over domain knowledge
- Production inference engineering -- making research-grade ideas run reliably at scale with human oversight
From the Aigentic charter:
Human intelligence, amplified. The future of research lies not in replacing human expertise, but in augmenting it with advanced AI capabilities that accelerate discovery and enhance understanding.
The bottleneck in most organizations isn't compute -- it's the gap between what AI can theoretically do and what actually ships. I close that gap.
Building in public from 2026. Unknown, SWE.