About
AI systems architect. Independent researcher.
I architect AI systems built to be trusted with real capital.
Twenty years of building software systems — the last two and a half dedicated to ZVAKTHOR, an algorithmic execution platform for global financial markets.
Most agentic AI stops at a working demo. The hard part is everything after it: execution governance, observability, and failure handling. That is the part I build.
- Multi-agent orchestration and agent reliability
- Decision intelligence: turning unstructured context into auditable decisions
- Governance and risk architecture at the Python / Rust boundary
- Variance-based noise filtering and volatility estimation
ZVAKTHOR separates intelligence from control. Specialist agents analyse markets independently; an isolated Rust core validates every decision for risk and execution eligibility before any order reaches the market. Every decision, accepted or rejected, is recorded and replayable.
Designed for non-custodial licensing: clients would run the software on their own broker accounts, capital stays in their name, and the system never takes discretionary control.
Author of seven technical books on AI, blockchain, and quantitative trading, and four SSRN working papers on deterministic control architecture for agentic trading systems, staged-capital deployment, and signal-discovery design. M.Sc. Software Engineering.
Identifiers
- ORCID
- 0009-0009-4656-5983
- SSRN
- ssrn.com/author=13197688
- in/hadi-soufi-463170266
- GitHub
- github.com/hadisoufi20
- Amazon
- Author page
- hadi@hadisoufi.com