Skip to content
View HeshamFS's full-sized avatar
🏗️
building
🏗️
building

Highlights

  • Pro

Block or report HeshamFS

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

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

Report abuse
heshamfs/README.md
Typing SVG

Hey, I'm Hesham 👋

I build agentic AI systems, MCP tooling, and numerical simulation workflows that do more than autocomplete politely.
Currently at ICAMS · Ruhr-Universität Bochum, working where generative AI meets computational materials science.

Website LinkedIn X / Twitter

Agentic AI MCP Tooling Numerical Modeling Computational Materials Python TypeScript

Professionally translating “this should be automated” into production code.


agent.runtime.yaml

codename: HeshamFS
class: human_with_agentic_tendencies
location: Germany
role:
  - GenAI specialist
  - Numerical modeling expert
  - Builder of useful things that occasionally scare manual workflows
specialties:
  - multi-agent systems
  - MCP servers and tool orchestration
  - numerical simulation workflows
  - computational materials science
  - research-grade automation
current_mission:
  - make AI genuinely useful for engineering and science
  - give agents enough numerical common sense to stop breaking physics
alignment: chaotic-good-with-tests
known_issue:
  - gets suspicious when a solver converges too quickly
github contribution grid snake animation

fun_faq

Why multi-agent systems?
Because asking one model to do everything is the AI equivalent of staffing a research lab with one sleep-deprived intern and a motivational poster.
What kind of problems interest you most?
The ones where language meets tooling, automation meets reality, and the outcome matters outside the chat window.
What breaks your heart a little?
Agents without tools. Benchmarks without baselines. Simulation workflows held together by folklore and clipboard history.
What do you optimize for?
Systems that are useful, testable, extensible, and just autonomous enough to save real time without inventing new categories of chaos.

send_signal

Website LinkedIn X / Twitter


Built in Germany with coffee, constraints, and a healthy distrust of brittle workflows.

Pinned Loading

  1. coalitionkit coalitionkit Public

    An open kit for groups whose AI agents work toward one goal together, with every decision signed by a person.

    Python 1

  2. materials-simulation-skills materials-simulation-skills Public

    Agent Skills for computational materials science: numerical stability, solvers, meshing, convergence, V&V, HPC and simulation workflows.

    Python 93 7

  3. swarmweaver swarmweaver Public

    Autonomous coding agent with swarm orchestration—multiple agents in parallel worktrees. Six modes: greenfield, feature, refactor, fix, evolve, security. CLI + Web UI, approval gates.

    Python 4

  4. consilith consilith Public

    Multiplayer agentic workspace: people and their agents collaborate through owned inputs, approvals, recorded execution and signed evidence.

    Python 2

  5. mcp-tool-factory-ts mcp-tool-factory-ts Public

    Generate MCP servers from natural language, OpenAPI specs, or database schemas.

    TypeScript 1

  6. materials-mcp-commons materials-mcp-commons Public

    Plugin-agnostic materials-science contracts and engine infrastructure for MCP.

    Python 3