Intelligent systems and platforms for organisations of all types and sizes.
Aperintel is a technology and AI company that builds intelligent systems. We design and operate software that surfaces and structures knowledge, hands the repetitive work to machines and the judgement to people, and ties products together so teams are not left stitching point solutions by hand. The products span technology, healthcare, education, finance, community, and ops.
Where it matters, accountability is built into the substrate rather than bolted on at the end: cryptographic audit chains, guardrail evaluation, scope-deviation detection, and multi-provider resilience. The proof of what a system did is a property of the system, not a separate compliance product.
Products across various sectors: technology, healthcare, education, finance, community, and ops. Some surfaces are publicly available at their own domains. Others are in development or in active use with design partners. The unifying thread across all of them is provability and embedded accountability, not the vertical.
The flagship public surface is aperintel.com, which carries the Lumen streaming advisor, the intelligence maturity scorer, the analytical brief generator, and the project scoping workflow. The Aperintel AI Academy runs cohort-based engineering education on the same accountability-first principles that shape the rest of the family.
- Nexuscone: tamper-evident append-only audit ledger with SHA-256 hash chain, optional Ed25519 signing, and optional Bitcoin anchoring via OpenTimestamps. Apache 2.0 on PyPI. The substrate underneath the Aperintel governance product family.
- NovaBash: local-first developer secrets vault. API keys encrypted in the browser with AES-256-GCM under a passphrase, a 24-word recovery phrase, per-environment
.envgeneration, a hash-chained local audit log, and an installable PWA, with no backend. Apache 2.0, live at novabash.pages.dev. - Mini-Assistant-Swarm: fork-friendly personal subagent toolkit. VERIFIER-gated outputs, never sends on the operator's behalf, always writes to disk. Apache 2.0.
- awesome-ai-governance: curated list of resources for building, shipping, and auditing AI in regulated environments. Regulations, standards, OSS primitives, governance platforms, security tools. CC0.
- regulated-ai-checklist: practitioner pre-deployment checklist for shipping AI in regulated environments. Governance, audit, security, data protection, monitoring, incident response, supply chain. CC BY 4.0.
More open-source projects are on the way, including the shared landing template used across the product family and a glossary aligned to the EU AI Act vocabulary.
The Aperintel Accountability Architecture Series is a programme of research on cryptographic accountability primitives for AI systems. Papers deposit to Zenodo, SSRN, and OSF, and anchor the engineering work to the academic literature.
Sponsor Aperintel on GitHub to support the open-source primitives that underpin the commercial products. Sponsorship funds maintenance of the Apache 2.0 substrate.
Design partner enquiries, customer conversations, and academic collaboration: enquiries@aperintel.com.
Security disclosure: open a GitHub Security Advisory on the affected repo, or email enquiries@aperintel.com with the subject line [<product> Security].
Aperintel is based in London. The work is one part of a larger thesis. As AI moves from helpful assistant to operational system, the systems that succeed will be the ones that can show their work.