Skip to content
View Tarunjit45's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report Tarunjit45

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.

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
Tarunjit45/README.md

⚔️ TARUNJIT BISWAS

AI PRODUCT ENGINEER · BUILDER · AI-ASSISTED DEVELOPER

╔══════════════════════════════════════════════════════════════╗
║                                                              ║
║              「 THE SYSTEM HAS BEEN ACTIVATED 」             ║
║                                                              ║
║                   PLAYER: TARUNJIT                          ║
║                   CLASS: AI PRODUCT ENGINEER                 ║
║                   RANK: S → ???                              ║
║                                                              ║
║        BUILD  →  BREAK  →  LEARN  →  EVOLVE  →  SHIP       ║
║                                                              ║
╚══════════════════════════════════════════════════════════════╝


🩸 「 WHO AM I? 」

I don't wait for the perfect idea.

I find a problem.
        ↓
Understand it.
        ↓
Design the system.
        ↓
Use AI as my multiplier.
        ↓
Build the product.
        ↓
Break it.
        ↓
Fix it.
        ↓
Deploy it.
        ↓
Level up.

I'm a B.Tech CSE student and AI Product Engineer focused on building AI-powered products, intelligent agents, RAG systems, developer tools, automation platforms and full-stack applications.

I enjoy taking an idea from:

"What if...""Let's build it.""It's live."

My strongest weapon isn't typing code faster.

It's turning ambiguous ideas into working systems.


⚡ THE SHINOBI PATH

A shinobi is not defined by the techniques they know.

They're defined by how quickly they adapt.

                    ┌──────────────────┐
                    │     UNKNOWN      │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │    EXPLORE       │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │     BUILD        │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │     BREAK        │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │     LEARN        │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │     EVOLVE       │
                    └────────┬─────────┘
                             ↓
                    ┌──────────────────┐
                    │      SHIP        │
                    └──────────────────┘

🥷 My Developer Philosophy

Learn → Build → Fail → Understand → Rebuild → Ship

I don't believe in waiting until I know everything.

I learn what the problem demands, build the system, discover what I don't know, and then go deeper.


🎴 CURRENT CHARACTER STATS

╭────────────────────────────────────────────────────────────╮
│                    PLAYER PROFILE                           │
├────────────────────────────────────────────────────────────┤
│                                                            │
│  🧠 Intelligence        ███████████████░░░  BUILDING       │
│  ⚔️ Engineering         █████████████░░░░░  LEVELING       │
│  🤖 AI Systems          ███████████████░░░  ACTIVE         │
│  🏗️ Product Building    ████████████████░░  STRONG         │
│  🔥 Curiosity           ██████████████████  MAX            │
│  🧪 Experimentation     █████████████████░  HIGH           │
│  📚 DSA                 ███████░░░░░░░░░░░  TRAINING       │
│                                                            │
╰────────────────────────────────────────────────────────────╯

🗡️ MY ARSENAL

🤖 AI / LLM

LLMs RAG AI Agents LangChain LangGraph Prompt Engineering LLM Evaluation Vector Search Embeddings AI Automation Local LLMs

⚔️ Engineering

Python TypeScript JavaScript FastAPI Node.js Express React Next.js REST APIs

🏰 Infrastructure

Docker PostgreSQL Redis pgvector MongoDB Firebase Git GitHub CI/CD

🧠 AI DEVELOPMENT

ChatGPT Claude Gemini AI-assisted coding AI-assisted debugging rapid prototyping agentic development

AI is not my replacement for engineering.

It's my multiplier.


⚔️ THE TITANS I'M BUILDING AGAINST

┌──────────────────────────────────────────────────────────┐
│                                                          │
│   🧩 COMPLEX SYSTEMS                                     │
│                                                          │
│   🤖 Autonomous Agents                                   │
│   🧠 LLM Applications                                    │
│   🔎 Retrieval-Augmented Generation                      │
│   🛡️ AI Safety & Evaluation                              │
│   ⚙️ Developer Infrastructure                            │
│   🏗️ Full-Stack AI Products                              │
│   🚀 Autonomous Software Engineering                     │
│                                                          │
└──────────────────────────────────────────────────────────┘

I don't want to build another chatbot.

I want to build systems that do things.


🏯 FLAGSHIP PROJECTS

⚔️ Autonomous SWE Agent

An AI software engineer that attacks GitHub issues.

GitHub Issue
     ↓
Repository Analysis
     ↓
Relevant Code Retrieval
     ↓
Reasoning / Planning
     ↓
Code Generation
     ↓
Sandbox Execution
     ↓
Tests
     ↓
Failure Analysis
     ↓
Iterative Fix
     ↓
Pull Request

Stack: Python LangGraph FastAPI ChromaDB GitHub API Docker

Focus: Autonomous coding · Repository intelligence · Agent orchestration · Evaluation


🏯 Conflux AI

AI infrastructure for businesses.

A multi-tenant AI platform combining intelligent business workflows, document intelligence, RAG, automation and conversational AI.

                    CONFLUX AI

                     USER
                       │
                       ▼
                ┌──────────────┐
                │ AI INTERFACE │
                └──────┬───────┘
                       │
        ┌──────────────┼──────────────┐
        ↓              ↓              ↓
     RAG          AUTOMATION      BUSINESS AI
        │              │              │
        └──────────────┼──────────────┘
                       ↓
                ┌──────────────┐
                │  DATA LAYER  │
                └──────────────┘

Stack: Next.js FastAPI PostgreSQL pgvector Redis Docker

Focus: SaaS · Multi-tenancy · RAG · AI workflows · Production architecture


👁️ AgentWatch

Observability for AI agents.

Monitor what your agents are doing.

Agent
  │
  ├── Events
  ├── Tokens
  ├── Latency
  ├── Cost
  ├── Errors
  └── Tool Calls
         │
         ▼
   ┌─────────────┐
   │ AgentWatch  │
   └──────┬──────┘
          ↓
   Real-time Dashboard

Stack: FastAPI PostgreSQL Redis WebSockets Textual

Focus: AI observability · Telemetry · Cost tracking · Real-time systems


🧪 EXPERIMENT LAB

I'm constantly experimenting with:

┌────────────────────────────────────────────┐
│                                            │
│  🧠 LLM Applications                       │
│  🤖 Autonomous Agents                      │
│  🔎 RAG & Retrieval                         │
│  🛡️ AI Safety                              │
│  👁️ Computer Vision                        │
│  ⚙️ Developer Tools                        │
│  🧬 Multi-Agent Systems                     │
│  🚀 Autonomous Software Engineering        │
│                                            │
└────────────────────────────────────────────┘

Some experiments become products.

Some become failures.

Both are useful.


📈 THE LEVELING SYSTEM

                    CURRENT QUEST
                         │
                         ▼
              ┌─────────────────────┐
              │ AI PRODUCT ENGINEER │
              └──────────┬──────────┘
                         │
             ┌───────────┼───────────┐
             ↓           ↓           ↓
          BUILD       DEPLOY      LEARN
             │           │           │
             └───────────┼───────────┘
                         ↓
                    NEXT LEVEL
                         │
                         ▼
                AI SYSTEMS ENGINEER
                         │
                         ▼
                 AI RESEARCH ENGINEER
                         │
                         ▼
                    ??? / AGI

🎯 Current Quests

  • Build production-grade AI systems
  • Contribute meaningfully to open source
  • Work with an exceptional engineering team
  • Go deeper into AI systems and infrastructure
  • Build systems used by real people
  • Become significantly better at software engineering
  • Explore autonomous intelligence
  • Eventually work on frontier AI

🪽 SCOUT REGIMENT MINDSET

"If we don't fight, we can't win."

The unknown doesn't scare me.

Complex systems don't scare me.

A technology I don't know doesn't scare me.

The only thing that bothers me is staying at the same level.

So I keep exploring outside the walls.

                    THE WALLS
        ┌─────────────────────────────┐
        │                             │
        │       KNOWN TECHNOLOGY      │
        │                             │
        │       KNOWN SYSTEMS         │
        │                             │
        └──────────────┬──────────────┘
                       │
                       ▼
                 UNKNOWN TERRITORY

              AI × SYSTEMS × AGENTS
                       ×
                AUTONOMOUS SOFTWARE
                       ×
                  ??? / AGI

🧩 OPEN SOURCE

I believe the fastest way to become a better engineer is to:

Read → Understand → Modify → Break → Fix → Contribute

I'm interested in contributing to projects around:

  • AI agents
  • LLM infrastructure
  • developer tooling
  • AI evaluation
  • RAG
  • autonomous software engineering
  • AI safety

📊 GITHUB INTELLIGENCE


🐉 THE ROAD AHEAD

             ┌───────────────────────┐
             │       BEGINNER        │
             └───────────┬───────────┘
                         ↓
                   BUILD PRODUCTS
                         ↓
                   SHIP SYSTEMS
                         ↓
                 UNDERSTAND SYSTEMS
                         ↓
                  MASTER AI SYSTEMS
                         ↓
                  BUILD AUTONOMOUS
                     SYSTEMS
                         ↓
                         ?

I don't know exactly where the road ends.

That's the point.


⚡ IF YOU'RE HERE...

If you're an engineer, researcher, founder, builder, or someone obsessed with AI and technology:

Let's build something difficult.

┌─────────────────────────────────────────────┐
│                                             │
│          「 THE NEXT LEVEL AWAITS 」         │
│                                             │
│              BUILD SOMETHING                │
│                IMPOSSIBLE.                  │
│                                             │
└─────────────────────────────────────────────┘

⚔️ BUILD · 🧠 LEARN · 🤖 AUTOMATE · 🚀 SHIP · 🔥 EVOLVE


Pinned Loading

  1. ChainWatch ChainWatch Public

    ChainWatch is a flight data recorder for multi-step AI systems. It's a CLI-based tool that records every step in an AI decision chain, links them together in order, prevents tampering, and allows y…

    Python 1

  2. SpecGuard SpecGuard Public

    SpecGuard is a command-line tool that turns AI safety policies and behavioral guidelines into executable tests. Think of it as unit testing for your AI's output. Instead of trusting that your AI wi…

    Python 1

  3. ModelPulse ModelPulse Public

    ModelPulse helps maintain model reliability and performance by providing early warning signals for these issues, allowing teams to address them before they impact users significantly.

    Python 1