Skip to content
View fswair's full-sized avatar
🤠
coding
🤠
coding
74 followers · 151 following

Achievements

Achievement: StarstruckAchievement: Pull Sharkx2Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Public Sponsor

Achievements

Achievement: StarstruckAchievement: Pull Sharkx2Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Public Sponsor

Highlights

Organizations

@vcoderun

Block or report fswair

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. Markdown is supported. 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
fswair/README.md

Hello there 👋

I'm Mert, a Harness Engineer and OSS Maintainer focused on Agentic Systems & Applied AI.

I specialize in agent harnesses, multi-agent orchestration, CodeMode, and agent interoperability. I work with A2A, ACP, MCP, Pydantic AI, LangChain, Codex App Server, and Claude Agent SDK, with a focus on typed APIs, evaluation, and observability.

fswair

Website · LinkedIn · PyPI · Email

Languages & Tools

Area Stack
Languages Python TypeScript Modern C++ SQL
Python & backend Pydantic FastAPI PyPI
Agent frameworks & runtimes Pydantic AI Pydantic Monty LangChain Codex App Server Claude Agent SDK
AI Models Decision Models Large Language Models
Optimization & evaluation DSPy GEPA Pydantic Evals LLM-as-a-Judge
Testing & observability Logfire OpenTelemetry pytest Codecov GitHub Actions
Protocols A2A ACP MCP OTLP LSP
Development tools Zed Codex Tree-sitter
Infrastructure & data Docker AWS NGINX SQLite MongoDB

Projects

Project What it does
VSH Rust-powered transactional filesystem simulation: execute in a virtual snapshot, inspect changes, and commit through explicit policies. Docs
AutoBench Benchmarking infrastructure for comparing variants, capturing semantic evidence and asset versions, and replaying recorded experiments. Docs
ACP Kit ACP adapters for Pydantic AI and LangChain / LangGraph, with session state, approvals, runtime capability mapping, and remote transport helpers. Docs
pydantic-gepa A typed optimization runtime combining GEPA, Pydantic AI, and Pydantic Evals, with candidate injection, reflection, and checkpoint/resume. Docs

GitHub Activity

GitHub statistics for fswair: commits, stars, pull requests, and issues GitHub contribution streak for fswair

Interested in collab about agentic systems, harnesses, or evaluation? contact@tomris.dev

Pinned Loading

  1. vcoderun/autobench vcoderun/autobench Public

    YAML-first benchmark and evidence framework

    Python

  2. vcoderun/acpkit vcoderun/acpkit Public

    ACP Kit provides a common adapter for Agent Frameworks.

    Python 22 3

  3. pydantic/pydantic-ai pydantic/pydantic-ai Public

    How Python does AI. Agents, realtime voice, image generation, embeddings. Every model, every interface, typed end to end.

    Python 20.5k 2.9k

  4. vsh vsh Public

    Low-latency filesystem simulation for AI agents operating on workspaces.

    Rust 1

  5. vcoderun/vcode vcoderun/vcode Public

    The shim to use agent that is type safe & fully tested for frontier coding.

    Python

  6. vcoderun/vowel vcoderun/vowel Public

    YAML Based Eval Specification Language and AI generation pipeline for LLMs and Developers.

    Python