Sepehr Mahmoudian
I build domain-specific AI agents and custom harnesses for auditable workflows, alongside AI software for LLM/VLM evaluation and traceability, knowledge graphs and Graph RAG. My research spans Partial Information Decomposition (PID), information-theoretic representation analysis, computational neuroscience, robotics and multimodal 3D perception. Mostly Rust and Python.
Focus
Selected work
- engram : Engram is a research-stage, domain-specific AI agent with a custom harness for biophysical and functional neural models. It orchestrates provenance-aware, closed-loop computational-neuroscience simulations for reproducible R&D. The public repository remains a placeholder pending source publication; it does not claim general paper reproduction or validated scientific results.
- NCP : NCP is a versioned, project-agnostic canonical-JSON contract connecting neural simulators or neuromorphic controllers to robots, UAVs, simulators and read-only analysis clients. It has a Rust reference, an independent TypeScript validator/client and Python/C/C++ bindings; HEAD is an unreleased, release-blocked 1.0 candidate, while v0.8.0 remains the latest immutable release.
- prisoma : Prisoma is a public Rust/Python research toolkit implementing canonical experiment run logs and replay groundwork while working toward provenance-complete capture-intervention-replay for intervention-grounded VLA-policy diagnosis. Real VLA capture and complete replay acceptance remain blocked; PID is a conditional diagnostic behind preregistered validity gates. The public 0.9.0 GitHub source prerelease is a research-software preview, not a scientific-results or production release.
- crebain : CREBAIN (Adaptive Response & Awareness System) is a research-only spatial-awareness prototype that combines Gaussian-splat visualization and simulated cameras with ML detection, persistent multimodal 3D tracking and ROS/Gazebo drone simulation. Drone-physics and ROS capabilities remain in progress.
- melkor : Melkor provides deterministic 3D Gaussian Splatting (3DGS) conversion across PLY, SPZ and glTF. Its inspection reports surface field provenance, bounds and numeric hazards without modifying source assets. The current release candidate is source-only; no production binary is currently supported.
- galadriel : Galadriel is an experimental safe-Rust research prototype for cross-sensor consistency monitoring, combining NIS/CUSUM, sign-preserving correlation over producer-attested projections and optional PID diagnostics. The 0.9.0 research-source target remains unpublished; evidence is synthetic/component-level, not field validation.
- pid-rs : pid-rs is a safe-Rust library for shared-exclusions Partial Information Decomposition and mutual-information estimation: categorical SxPID, KSG MI and default-off experimental continuous shared-exclusions/PID surfaces. v0.9.0 is a GitHub-only source-review prerelease.
- haldir : Haldir Gate is an experimental Rust mission-authorization reference monitor for NCP: it validates signed controller intents, deployment and mission admission, and deterministic policy, then prepares Gate-owned plant commands and signed decision receipts. Its 0.9 review remains NO_GO; it is not production-ready or airworthy.
- manwe : Manwe is an alpha airspace-perception research and validation workbench spanning vision, audio, multi-camera geometry and multi-target tracking, with a Python numerical/training package and Rust/Candle inference benchmarks. It targets candidate outputs for systems such as Crebain but currently ships no drop-in Crebain adapter.
- cortexel : Cortexel is an unreleased TypeScript library and CLI for neural-simulation figures: it validates and canonicalizes strict declarative JSON requests with fail-closed provenance, then returns deterministic SVG plus a complete exact-value table. main is 0.10.0-dev.0; v0.9.0 is the last tagged preview.
- relief-atlas : relief-atlas is a Python generation pipeline and 10,079-item manifest and prompt catalog for disaster-relief, humanitarian-aid and civil-protection meshes. The repository currently contains 125 GLBs, not a complete 10,079-mesh corpus.
- cobot-atlas : cobot-atlas is a Python generation pipeline for a public MIT dataset of 2,023 unique glTF 2.0 Binary meshes (2,024 GLB files, 33.5 GB) for robot/cobot simulation, manipulation research, VLA training and benchmarking. Hugging Face dataset · Dataset DOI · Pipeline DOI
Writing & research
- [Re] Measures for investigating the contextual modulation of information transmission : ReScience C 6(3), 2020 · reproducibility repository.
- Google Scholar : publications.
- Substack : notes & essays.
FAQ
Who is Sepehr Mahmoudian?
A Senior AI Engineer based in Berlin, Germany, with more than a decade of experience across applied AI, research software and computational neuroscience. He builds domain-specific AI agents and custom harnesses, and works across LLM/VLM evaluation and traceability, knowledge graphs, Graph RAG, robotics and multimodal 3D perception. GitHub member since 2014.
What does he work on?
Domain-specific AI agents and custom harnesses; LLM/VLM evaluation, traceability and observability; knowledge graphs and Graph RAG; information-theoretic representation analysis and Partial Information Decomposition; robotics and multimodal 3D perception, primarily in Rust and Python.
What is his current research?
His PhD dissertation is in final preparation. Its core develops new mathematical PID analyses and information-theoretic methods for studying neural representations. A supporting research-software ecosystem includes Engram, a research-stage, domain-specific AI agent and custom harness for provenance-aware, closed-loop simulations of biophysical and functional neural models, alongside validation, perception and control tools for reproducible, auditable computational-neuroscience R&D.
What are his main projects?
His selected work spans computational neuroscience, neural-control protocols, intervention-grounded VLA-policy diagnosis, spatial awareness and airspace perception, statistical consistency and authorization monitors, 3D Gaussian-splat tooling, deterministic scientific visualization and synthetic 3D datasets. Each project above links to its public repository and evidence.
Has he published research?
Yes, including the ReScience C replication [Re] Measures for investigating the contextual modulation of information transmission. Publications are listed on Google Scholar.
Where is he based and how can he be reached?
Berlin, Germany. By email at sepmhn@gmail.com, or via the links below.
Curriculum vitae
Public three-page editions in English and German.