Skip to content
View CodeWithJuber's full-sized avatar
🎯
Focusing
🎯
Focusing
29 followers · 227 following

Achievements

Achievement: Pull Sharkx4Achievement: QuickdrawAchievement: Pair Extraordinairex2Achievement: YOLOAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pull Sharkx4Achievement: QuickdrawAchievement: Pair Extraordinairex2Achievement: YOLOAchievement: Arctic Code Vault Contributor

Block or report CodeWithJuber

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
CodeWithJuber/README.md

Typing SVG

Juber Shaikh

Senior Full-Stack & AI Engineer · Dubai, UAE (UTC+4) · Open to fully-remote roles & contracts

11+ years building production systems across Java/Spring Boot, Python, React, and AWS — now focused on AI systems: agent reliability infrastructure, governed connectors, and durable agent memory. I also publish independent research across AI/NLP, maths, and philosophy of physics.

Java Spring Boot Python React Next.js TypeScript Rust AWS Docker Kubernetes

🔧 What I build

Project What it is Status
forgekit Reliability infrastructure for AI coding agents — evidence-linked memory, change-impact analysis, guardrails, cost governance. Multi-command CLI + MCP server, zero runtime deps. Emits native config for Claude Code, Codex, Cursor, Gemini, Aider, Copilot, Windsurf, Zed, Continue, OpenClaw. v1.9.0 · MIT · CI + CodeQL + Scorecard green
mizan Evidence-first agentic AI research prototype — agent loop, provider-normalized tool calling, local memory, plugins. Beta · Apache-2.0
hikmah-stack Deterministic Rust memory & reliability kernel for AI agents — provenance-bearing memory, contradiction detection, bounded recall, decision controls. MIT
wisdomlens 4 thinking-quality skills for AI coding tools (Claude Code, Codex, Kimi, OpenClaw). MIT

🔬 Research

Independent researcher working across AI/NLP, mathematics, physics, and philosophy — not one narrow lane.

  • Q-CSMP v2 — a 28,000-row dataset for Qur'anic Arabic contextual sense disambiguation. Published open: Zenodo (DOI 10.5281/zenodo.23024527) · Hugging Face · CC-BY-4.0 · listed as a work on my ORCID record.
  • ΔA experiment (v2) — confirmatory run on 17,424 cases: ΔA = +0.0338, 95% CI [0.0025, 0.0648] — weakly supports H1, with a noted overconfidence caveat. Negative results reported too (Stage-2 ΔQ pilot: −0.0555).
  • Prime-number research — ongoing work on prime gaps and formula-based results.
  • Papers — two 8-page ACL-format drafts (dataset paper + experiment paper), currently unpublished drafts.

🎓 Academic

  • B.E. Electronics & Telecommunication, University of Mumbai — 2014, First Class.
  • ORCID: 0009-0009-3546-3455 (verified) · research datasets on Zenodo and Hugging Face.
  • Position: Independent Researcher — no institutional affiliation; all work self-directed and openly published.

💼 Open to work

  • Fully-remote engineering roles and contracts — software, AI, and research-adjacent work.
  • Rate: USD 100/hr for contract engagements.
  • Stack: Java/Spring Boot · Python · React/Next.js · Node.js · Rust · AWS · Docker/Kubernetes.
  • 🌐 jubershaikh.com · 💼 LinkedIn · 📧 Juber@hostlelo.com

📊 GitHub stats

Juber's GitHub stats Top Langs

Every claim above is verifiable in the linked repos and records. The evidence is the code, the CI, and the data — nothing inflated.

Pinned Loading

  1. forgekit forgekit Public

    Reliability layer for AI coding agents: shared memory that survives sessions, blast-radius checks before edits, verification gates after — one config for 10 tools, zero runtime deps.

    JavaScript 2

  2. hikmah-stack hikmah-stack Public

    Deterministic Rust memory and reliability kernel for AI agents — provenance-bearing memory, contradiction detection, bounded recall, commitments, decision controls.

    Rust 1

  3. wisdomlens wisdomlens Public

    An operating layer for your agent's judgment — instruction-only skills that make it verify before it acts, specify before it builds, and ship artifacts instead of foam. 4 skills, an advisor agent, …

    JavaScript 1

  4. Host-Lelo/roadlimit-uae Host-Lelo/roadlimit-uae Public

    Privacy-first open-source speed awareness for Android/iPhone — Dubai-first research beta (React Native/Expo). Not store-ready; not a government service.

    TypeScript 1

  5. connector-hub connector-hub Public

    Omni Connector Hub — one channel for AI providers (OpenAI/Claude/Kimi/Cloudflare), multi-Gmail OAuth, WHMCS/WHM/cPanel, VPS clouds (Hetzner/Linode/Contabo/OVH/OneProvider/UltaHost), tawk.to, GitHub…

    Rust