AI ENGINEER — BUILDER OF BROK'S FORGE

Engineering AI that survives production.

demos are easy. surviving is the job.

Evaluation pipelines, RAG systems, agent registries, developer tooling — designed, built and operated end to end. The flagship is Brok’s Forge: an open-source engineering platform for AI agents.

  • Software Engineer @ TCS
  • Google Cloud · BigQuery
  • 2 systems live in production

SHEET 01 — THE LOOP

runs on every merge
AgentEvaluateBenchmarkAdvise

94.2%

pass rate

1.8s

p95 latency

4,500

results scored

now building: async evaluation workers

drawingevaluation loop
revv1.0.0
scaleproduction
drawn byg.m.

fig. 01 — one evaluation job: 500 items fan out, 9 metrics score every run.

  • 22backend modulesone platform, clean boundaries
  • 29append-only migrationsthe schema is the contract
  • 13LLM providers modeledbehind one SPI
  • 2systems in productionbuilt and operated solo
Flagship

Brok's Forge — The Engineering Platform for AI Agents.

Frameworks help you build an agent. Brok’s Forge is the platform that helps you engineer one — register it, version it, point real datasets and prompts at it, evaluate it against objective metrics, benchmark variants, catch regressions, and report on cost, latency and quality over time.

the résumé is the repo →
Brok's Forgev1.0.0 · Apache 2.0
Live in production

The engineering loop it owns

  1. Register
  2. Version
  3. Evaluate
  4. Benchmark
  5. Advise

Framework-agnostic (LangGraph, CrewAI, AutoGen, Spring AI, custom HTTP) and provider-agnostic (OpenAI, Anthropic, Gemini, Groq, Ollama and more) — every provider behind one SPI, every result reproducible by construction.

  • 22feature & infra modules
  • 27REST controllers
  • 29append-only migrations
  • ~335backend Java files
  • 13LLM providers modeled
  • 17architecture decision records
Also in production

Kuriosity — the other side of the loop.

Brok's Forge evaluates AI systems; Kuriosity is one — a live RAG product I built and operate end to end. Building both is the point: I understand the platform and the workload it serves.

Kuriosity
Live in production

Kuriosity inverts the AI-tutor model: an AI named Kurio plays the student. Upload study material and Kurio asks context-grounded questions, scores your free-text answers against the retrieved source chunks, and tracks topic mastery across sessions.

Custom RAG pipeline

PDF parsing, chunking, Cohere embeddings, ChromaDB similarity search — every response grounded in the uploaded source, not the model’s general knowledge.

LLM evaluation loop

Free-text answers scored against retrieved chunks — accuracy %, missing concepts and revision topics returned as structured JSON.

Cross-session memory

PostgreSQL tracks per-user topic strength, so weak concepts resurface until they hold.

Operated in production

JWT auth with email OTP, rate limiting, and a three-platform deploy (Vercel, Railway, custom DNS/TLS) run under real traffic.

  • React
  • Node.js
  • Express
  • PostgreSQL
  • ChromaDB
  • Cohere
  • Groq
Earlier systems

Applied computer vision and systems fundamentals — the range behind the platform work.

All repositories
Applied computer vision

Violence Detection & Alert System

Real-time CV pipeline that detects violent activity in video streams and triggers automated safety alerts — YOLOv6 + MobileNetV2 over live footage.

  • Python
  • YOLOv6
  • MobileNetV2
Applied computer vision

Road Safety Surveillance

Detection and tracking of road-safety violations in real time, replacing constant human monitoring with an automated pipeline.

  • Python
  • OpenCV
  • ML
Systems fundamentals

Student Management System

A clean object-oriented CRUD application for student records — early proof of the same instinct: model the domain first.

  • Java
  • OOP
System overview

The full production surface.

Everything below has shipped in real systems — the platform, the products, or the day job. No aspirational logos.

no fluff. counted twice.

Backend & runtime

  • Java 21
  • Spring Boot
  • Spring Security
  • PostgreSQL
  • Redis
  • Flyway
  • REST / OpenAPI
  • Node.js

AI systems

  • LLM APIs (OpenAI · Anthropic · Gemini · Groq · Ollama)
  • RAG pipelines
  • embeddings & vector search
  • evaluation harness
  • prompt / dataset versioning
  • LLM-as-judge metrics
  • LangGraph · LangChain (integration)

Infrastructure & ops

  • Docker
  • AWS EC2
  • Nginx + TLS
  • Prometheus
  • Grafana
  • structured logging
  • health probes
  • OpenTelemetry-ready
  • CI/CD (GitHub Actions)

Data & cloud

  • GCP
  • BigQuery
  • Looker
  • LookML
  • SQL optimization
  • row-level security

Frontend

  • React
  • Next.js
  • TypeScript
  • Tailwind
  • Vite
  • TanStack Query

Practice

  • clean architecture
  • modular monolith
  • ADRs
  • schema-first migrations
  • multi-tenant security
  • SSRF / IDOR hardening
  • documented testing strategy
Operating principles

Principles I refuse to compromise.

I'm an AI engineer who treats AI systems as production software, not demos. By day I build governed cloud data platforms at Tata Consultancy Services; nights and weekends belong to the workshop — the registries, pipelines and evaluation loops that make AI agents shippable.

all learned the hard way
01

Platforms, not demos

Anyone can call a model API. The hard part is everything around it — versioning, credentials, datasets, evaluation, cost. I build that part.

02

Measured, not vibes

An agent that isn't evaluated is an agent you can't ship. Every system I build produces metrics, benchmarks and regression signals — quality is a number, not an impression.

03

The schema is the contract

Databases outlive code. Append-only migrations, immutable versions and reproducible results — so an evaluation run today means the same thing a year from now.

04

Boundaries over convenience

Modules reference each other by id and published services, never shared tables. It costs a query today and buys a clean extraction tomorrow.

05

Security by construction

Tenant isolation that returns 404 instead of leaking existence, encrypted credentials, SSRF guards on every outbound call. Defaults, not afterthoughts.

06

Built to be operated

Correlation IDs on every log line, structured errors, health probes, fail-fast startup. If an incident is hard to trace, the system is unfinished.

Build log

From the workshop.

On the bench now

Async evaluation workers — moving the executor seam behind a queue and a worker fleet.

the roadmap is public — hold me to it
  1. Jul 2026

    Production hardening pass on Brok’s Forge — provider-aware health checks, OTP password change, Prometheus metrics, structured JSON logging.

  2. Jul 2026

    Brok’s Forge v1.0.0 — all four phases live. 22 modules, 29 migrations, 17 ADRs, and a Master Architecture doc that is actually kept true.

  3. 2026

    Phase 4: the AI Engineering Advisor — recommendations, root-cause engine, AI debugger, knowledge graph. Everything computed on read, nothing stale.

  4. 2025

    Kuriosity shipped — a RAG tutor in production across three platforms. Still running.

Experience

The work behind the workshop.

Governed cloud data platforms by day; open-source AI infrastructure the rest of the time. Both are production systems with real users.

  1. Brok's Forge — Independent / Open Source

    2026 – Present

    Creator & Maintainer · AI Engineering Platform · RemoteCurrent

    • Designed and built a multi-tenant, provider-agnostic AI-agent engineering platform (Java 21 / Spring Boot, PostgreSQL, Next.js 15): agent registry, evaluation pipeline, benchmarking, regression detection and an on-read engineering advisor.
    • Architected it as a modular monolith — 22 modules with id-only boundaries — so any module extracts into a microservice mechanically.
    • Operate it in production on AWS EC2 (Docker, Nginx, Let’s Encrypt) with a Vercel frontend, Prometheus metrics and structured logging.
    • Java 21
    • Spring Boot
    • PostgreSQL
    • Next.js
    • AWS
  2. Tata Consultancy Services

    Dec 2025 – Present

    Software Engineer · Google Cloud / Looker · ChennaiCurrent

    • Design and develop production LookML models, views and explores on Google Cloud Looker, enabling self-serve analytics for business teams.
    • Optimized BigQuery SQL, reducing average dashboard load time by ~35%.
    • Built 10+ Looker dashboards across Finance, Operations and Marketing.
    • Implemented row-level security and access-group policies for multi-tenant data governance.
    • GCP
    • BigQuery
    • Looker
    • LookML
  3. NIT Trichy

    Sep 2024 – Oct 2024

    Game Developer Intern · Trichy

    • Shipped a full game on online-bullying awareness with Unity and C#, owning design through deployment.
    • Unity
    • C#
  4. MIC Chennai

    Aug 2023

    VR Developer Intern · Chennai

    • Built immersive VR experiences with Unity's XR Interaction Toolkit, improving simulation fidelity through iterative testing.
    • Unity XR
    • C#
  5. SRM Institute of Science & Technology

    2021 – 2025

    B.Tech, Computer Science Engineering · Trichy

Certifications

Credentials, quickly.

  • Google Cloud — Associate Cloud Engineer

    Google Cloud

  • Java (Certified)

    HackerRank

  • JavaScript (Certified)

    HackerRank

  • Python

    CodeChef

  • AR/VR Game Development

    CDAC Pune

Contact

Build something with me.

Open to AI engineering and platform roles — teams that care whether the system still works six months after the demo. If that's you, the inbox is open.

i read everything. usually same day.