The practitioner

One node in the loop.

Shivam SharmaSoftware Engineer · AI-Automation Architect · Technical Delivery

A worldview is only words until someone lives it. I'm one person trying to — an engineer by trade, building systems for a living, and using that same craft to draw the first pillar of this movement out of vision and into something real.

This is also, plainly, how I make my living and look for the work that funds the rest. So here is the practitioner behind the blueprint: what I do, what I've built, and the full résumé if you're hiring. Meet the engineer through the idea — and if it resonates, become a node yourself.


In brief

Software engineer with 10+ years spanning frontend engineering (Vue 3, React, TypeScript; PHP roots), AI-driven automation and systems architecture, and technical delivery. Designs and ships end-to-end systems — from drag-and-drop product builders to LLM-assisted validation pipelines — with a bias for measurable outcomes, evidence-led decisions, and human-in-the-loop reliability.

Frontend

Vue 3, React, TypeScript, JavaScript (ES2023), Pinia, Vue Router, Vite, Vuetify, HTML5, CSS/SASS

AI / Automation

LLM orchestration (Claude CLI), agentic pipelines & MCP tooling, closed-loop self-verifying agents, Playwright test generation, Figma→code automation, self-hosted LLMs (Ollama, Qwen 3), prompt systems, RAG / vector memory (Qdrant)

Backend / Systems

Node.js, Python (FastAPI), PHP, REST APIs, PostgreSQL, Redis/BullMQ, S3, queue/worker architecture, load-balanced multi-instance design

Delivery / Practice

Flow & systems design, cross-functional coordination, Jira/Confluence, code-review governance, technical documentation, CI/CD (GitHub Actions)

Experience

AI-Automation & Systems Engineer

2025 – Present

Visions, Chandigarh · InsureTech client (UK)

  • Architected and built a two-phase automated product-validation pipeline that shifts insurance-rate validation left from engineering to product managers, catching data errors before they reach development.
  • Designed a server-side LLM first-pass reviewer (headless Claude) running a 5-checkpoint review over raw insurer files with image- and checkbox-aware PDF parsing — generating a plain-language report and executable test cases at ~$0.75 and ~5 minutes per product.
  • Built the developer-stage automation: generated Playwright tests from raw rate/benefit files, ran them against the live quoting engine, and versioned certified test cases in S3 as the QA contract.
  • Quantified impact via a retroactive pilot across 7 production tickets — up to ~58% of historical dev↔PM clarification cycles automatically catchable — with bit-exact rate derivation validated on real insurer data and zero false positives across 4 shipped bundles after calibration.
  • Hardened the pipeline for horizontal scale: shared BullMQ/Redis queue (exactly-once execution across a load-balanced fleet), per-product idempotency, and surgical cross-instance job cancellation — verified live with zero duplicate or orphaned processes.
  • Conceived and ran an automated Figma → Locofy → GitHub → AI → Pull Request workflow: 51 automated design-to-PR runs across 3 modules in ~6 weeks, ~8 minutes per page, preserving business logic and coding standards.
  • Evolved that pipeline into a closed-loop, self-verifying design-to-code agent: a single `/locofy <ticket>` command reads a Jira ticket (Atlassian MCP), pulls the live Figma via the Locofy MCP, implements into React/TypeScript under an 'architect' rule, then self-verifies in a rendered Playwright loop (visual + exact-token diff) before opening a PR — validated end-to-end, targeting ~2–3 dev-hours/page vs ~12–16 hand-coded.

Senior Frontend Engineer (Vue 3 / TypeScript)

Oct 2023 – Aug 2025

Visions, Chandigarh · Corporate Learning & Performance Platform

  • #1 contributor of 30+ engineers (1,526 commits, +115k net lines over ~22 months) on a Vue 3 / TypeScript corporate Learning & Performance-Management platform.
  • Architected a drag-and-drop course-authoring tool (19+ content/question section types, content-JSON persistence) and a custom rich-text table editor with floating formatting controls.
  • Engineered a SCORM player and a SCORM-2004 package publisher from scratch (iframe sandboxing, zip parsing, package generation) — no vendor LMS runtime.
  • Built a multi-step CSV data-import pipeline with reusable column-mapping templates and alias resolution; implemented auth/MFA, forced password reset, and user impersonation.
  • Drove performance fixes (duplicate-request elimination, abortable API calls) and maintained code across modules authored by other engineers, resolving 400+ bugs.

Frontend Software Developer (React / Angular / Vue)

May 2019 – 2023

Visions, Chandigarh

  • Led front-end on a Qatar higher-education registration platform (HBKU): built a schema-driven dynamic-form renderer supporting 15+ field types and solely authored a conditional page-routing rules engine (all/any logic, type-aware operands) — top contributor across ~1,200 commits in three repositories.
  • Owned the interactive Google Maps discovery flow on an events, tours & experiences platform (Angular Universal SSR + NgRx + PWA): idle-driven search, in-flight request cancellation to eliminate call storms, and custom category map pins — 3rd of 28 contributors.
  • Led frontend delivery for large-scale client projects across React, Angular, and Vue — integrating REST APIs and real-time data handling, with SSR and hydration for faster loads and stronger SEO.
  • Partnered with project managers and clients on requirements, estimation, and milestone planning; mentored junior developers and ran code reviews to uphold modern coding standards.
  • Additional client work: Emerson & AD Ports ERP dashboards (Angular / Vue).

Beyond the work

  • Author — “The Vyas Matrix: Rewiring Your Brain for Power, Peace, and Ancestral Purpose.”

  • Leadership Potential Award (2022) and Performer of the Year (2021) — for guiding cross-functional teams and exceptional contributions to client success.

  • Built a self-hosted, air-gapped AI coding agent (Qwen 3 · Ollama · OpenHands · Qdrant) running at $0/request on a single GPU — documented as a developer build guide.

  • Self-directed study: South Asia–focused security studies (ongoing).

  • Ongoing research series — “The Cosmic Blueprint,” decoding the science beneath dharma and human unity. Read →

  • Languages: English, Hindi, Punjabi.