Seniya Mendis
Work

The systems, and how each one was built.

AI environments, developer tools, native frameworks, performance systems, and distributed infrastructure built at Senon Solutions.

About me
AI systemsIn production

VERONICA

Private AI operating environment

Cognition, voice, and computer use all execute on Senon Solutions' own model family, served through SGLang over its own network, so confidential material never reaches a third-party inference endpoint.

  • Python
  • Rust
  • TypeScript
  • SGLang
Read case study: VERONICA
FrameworkIn production since 2024

SenonUI

Native Python UI framework

Fine-grained reactivity for Python developers building native Windows applications: no Rust, no Dart, no JavaScript, and an engine written in Rust underneath that updates only what actually changed.

  • Python
  • Rust
  • System webview
  • WebGPU
Read case study: SenonUI
InfrastructureIn production

Senon Cloud

Global edge network and compute platform

An edge network across 120+ countries where client code runs natively at the edge in TypeScript, Python, and Rust, with a five-minute CPU ceiling per request.

  • Go
  • Rust
  • Erlang
  • WebRTC
Read case study: Senon Cloud
AI systemsIn production

Senon AI

In-house model family and orchestration layer

The model family behind every Senon Solutions AI system, built in-house so the company owns its own model repository and confidential work never has to reach someone else's.

  • Sparse MoE
  • Model fine-tuning
  • Multimodal AI
  • Agent orchestration
Read case study: Senon AI
Developer toolingIn development

Project O

Verified Rust acceleration for Python bottlenecks

Profiles a running Python application, finds the bottlenecks worth moving to Rust, generates the extension, and proves it behaves identically before anything is applied.

  • Rust
  • Python
  • PyO3
  • Maturin
Read case study: Project O
Developer toolingIn production

Senon Docs

Python API documentation with CI quality gates

Written in pure Python and pointed at any Python project, producing a full API reference in six formats, with coverage thresholds that fail a CI build when documentation quality regresses.

  • Pure Python
  • Typer
  • Rich
  • AST traversal
Read case study: Senon Docs
Desktop softwareIn production

Kameleon

Local-first universal file conversion

More than 3,100 conversion combinations across documents, images, audio, video, archives, and structured data, every one of them running on the machine, with nothing uploaded anywhere.

  • Rust
  • TypeScript
  • Local-first
  • File processing
Read case study: Kameleon
InfrastructureIn production

M-Core Autoscaler

ML-driven compute and GPU orchestration

An in-house alternative to Kubernetes that scales compute and GPU capacity from minimal TOML rather than sprawling YAML, with scaling decisions driven by machine learning instead of static thresholds.

  • Rust
  • Machine learning
  • GPU orchestration
  • Distributed systems
Read case study: M-Core Autoscaler
Systems engineeringIn production

M-Core Engine

Hardware-aware adaptive execution runtime

The core engine is written in Rust. It reads the machine at startup and derives every runtime budget from it, covering pools, caches, queue depth, connections, and sidecar policy, while verifying application integrity as it runs.

  • Rust
  • Python
  • TypeScript
  • NUMA
Read case study: M-Core Engine
Engineering range

From product interface to runtime and infrastructure.

Explore my expertise