Engineering domains
Deep technical work where
constraints shape the architecture.
Egothor projects are concentrated around four connected domains: retrieval, assurance, cryptography and the language/infrastructure layer beneath them.
Information Retrieval
Full-text search, indexing and result disclosure for environments in which authorization is part of the retrieval problem.
- Regulated and policy-aware search
- Deterministic indexing and query execution
- Information boundaries and disclosure
Software Assurance
Evidence-oriented engineering for test inventory, security classification and reproducible quality controls.
- Source-level test discovery
- Security-testing taxonomy
- Audit-ready evidence and CI outputs
Cryptographic Systems
Classical and post-quantum building blocks integrated into resilient data, signing and PKI workflows.
- Classical + post-quantum cryptography
- PKI and lifecycle engineering
- Offline and constrained-operation models
Language & Infrastructure
Compact algorithms and small infrastructure libraries intended to keep production paths deterministic and easy to reason about.
- Multilingual stemming
- Compiled trie transformations
- Shared context and event-driven coordination
Engineering posture
Research depth without research theatre.
Work is expected to make the underlying model visible: assumptions, state transitions, runtime constraints, tests and performance evidence should be concrete enough for another engineer to challenge.
Explicit models
Architecture should reveal the security, data and state model rather than leaving it implicit in incidental implementation behaviour.
Lean runtime paths
Preparation and validation can be complex when they make the production path smaller, faster and easier to reason about.
Evidence over assertion
Tests, benchmarks, deterministic artifacts and structured reports are part of the engineering model, not release decoration.