Skip to content
View francescomaiomascio's full-sized avatar
:atom:
thinking ICE...
:atom:
thinking ICE...
4 followers · 49 following

Block or report francescomaiomascio

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
francescomaiomascio/README.md

Francesco Maiomascio

LinkedIn     X     YouTube     Discord

Typing intro

Overview

I work on runtime architecture and execution models for long-running, stateful and cognitively-augmented software systems.

My focus is not on models, prompts, or benchmarks, but on
how systems execute, evolve, and remain governable over time.

I approach AI-enabled systems as software systems first: execution is explicit, authority is enforced at runtime level, and behavior remains observable, auditable, and replayable.

Research focus

  • Runtime-first system design
  • Execution governance & lifecycle control
  • Event-derived and replayable state
  • Capability-based side-effect management
  • Observability, traceability & audit
  • Agent-oriented execution environments

This work sits at the intersection of
systems engineering · distributed systems · AI infrastructure.

ICE — Intelligent Cognitive Ecosystem

ICE is an ongoing research and engineering effort exploring how to build:

  • explicit and governed execution environments
  • deterministic and replayable cognitive processes
  • memory as a validated, inspectable artifact
  • strict separation between inference and authority

ICE is not a framework and not a product.
It is a runtime architecture and execution model.

📘 Documentation & RFCs
https://francescomaiomascio.github.io/ice-docs/

https://github.com/francescomaiomascio/ice-docs

Execution philosophy

Models perform inference.
Runtimes decide what is allowed to happen.

Complexity is not something to hide.
It is something to structure, constrain, and observe.

My favorite tools & technologies


Python
C
Bash
Linux
Arch
Git

Docker
Kubernetes
Nginx
CI/CD
GitHub
VS Code

PostgreSQL
Redis
Kafka
GraphQL
Prometheus
Grafana

OpenTelemetry
Elasticsearch
RabbitMQ
gRPC
OpenAPI
Jupyter

GitHub activity

Support

This is independent, long-term research.

If you are interested in runtime-level system design, execution governance, and infrastructure for intelligent systems:

Sponsor

Buy Me a Coffee

Pinned Loading

  1. ice-foundation ice-foundation Public

    Normative and governance layer of the ICE ecosystem. Defines architectural laws, invariants, and RFCs.

  2. ice-docs ice-docs Public

    Official documentation for ICE — a runtime architecture for cognitive systems. Architecture, execution models, observability, and design principles for long-running intelligent systems.

  3. ice-runtime ice-runtime Public

    Execution core of the ICE ecosystem: lifecycle, events, capabilities, sessions, and runtime orchestration for intelligent systems.

    Python

  4. ice-studio ice-studio Public

    Development and experimentation environment for the ICE ecosystem, providing a graphical interface, plugins, and integration with ICE runtimes and agents.

    TypeScript

  5. ice-conscious ice-conscious Public

    Cognitive layer of the ICE ecosystem, modeling memory, knowledge, embeddings, and semantic context for awareness, retrieval, and reasoning.

    Python

  6. ice-api ice-api Public

    Public API and contract layer of the ICE ecosystem, defining schemas, actions, IPC messages, and interaction boundaries between systems.

    Python