Hi, I'm Yatin.
First-year B.Tech AI & Machine Learning student focused exclusively on Python Backend Engineering, building voice AI pipelines with Speech Recognition & ElevenLabs, multimodal models with Groq & PyTorch, and high-performance APIs with FastAPI, Flask & MySQL.
Magic happens behind the doors
Click any door to swing it open in 3D perspective. Click inside an open door to inspect that project's full architecture dossier and code.
JARVIS
SpeechRecognition, ElevenLabs Flash TTS, Groq AI, and voice typing computer automation.
Nova AI
Groq GPT-OSS-20B, Cloudflare FLUX 2 Klein, Magic Hour & local PyTorch MusicGen with CUDA.
Boutique Backend
Customer appointment booking API, dynamic price estimation, and MySQL schema init.
Backend & Vision
Python Core, FastAPI, OpenCV vision processing, PyTorch tensors & SQL schemas.
Backend Projects
JARVIS — Personal Voice Assistant
A Python-based personal AI assistant capable of understanding spoken commands, speaking natural responses, opening websites, controlling YouTube, answering general knowledge with Groq AI, executing Voice Typing, and maintaining conversation history.
$ python mega_project_jarvis.py
>> Initializing speech recognition & audio drivers...
>> ElevenLabs Flash Model: Connected
>> JARVIS is ready!
User: "Jarvis"
JARVIS: "Yes boss"
User: "Open YouTube and play Believer"
JARVIS: "Opening YouTube and playing Believer boss..."
User: "Start typing"
JARVIS: "Voice typing activated. Tell me what to type."
Nova AI — Multimodal Assistant
A Python-based multimodal AI assistant combining conversational intelligence with real-time web search, AI image generation/editing, AI video generation/animation, and locally running AI music generation.
generated_image.png).
$ python main.py
>> Loading MusicGen Small...
>> Device: cuda (NVIDIA GPU Acceleration active)
>> MusicGen loaded successfully!
Prompt: "Generate an image of a futuristic city"
Cloudflare: "FLUX 2 Klein rendered → generated_image.png"
Prompt: "Generate relaxing piano music"
PyTorch: "Synthesizing audio tensors → generated_music.wav"
Lord Krishna Boutique — Booking Engine
A Flask-based backend booking system for managing customer service reservations. Features automated database table creation, custom server-side input validation, and dynamic price calculation engine saved securely into MySQL.
.env.
init_db() creates and configures the bookings relational schema on server boot.
$ python app.py
>> Connecting to MySQL (lord_krishna_boutique)...
>> Table 'bookings' verified/initialized!
>> Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
POST /book: {"service": "Bridal Wear", "phone": "6280715453"}
Response: 200 OK — "Booking recorded. Notification sent!"
Year 2026: The AI Inception
First-Year B.Tech AI & ML journey dedicated to mastering Python backend architecture.
JARVIS Voice Assistant
Engineered SpeechRecognition, ElevenLabs Flash neural voice, Groq AI context, and voice typing PC control.
Voice & AutomationNova AI Multimodal Platform
Connected Cloudflare FLUX 2 Klein, Magic Hour AI video, and local MusicGen on NVIDIA CUDA GPU.
Generative AI & PyTorchLord Krishna Boutique Backend
Constructed Flask REST backend, automated MySQL schema creation, and server-side validation.
Flask & MySQLFastAPI & Vision Foundations
Mastered asynchronous FastAPI microservices, OpenCV computer vision frames, and basic PyTorch tensors.
FastAPI & OpenCVBackend Skills Arsenal
Python Core
OOP, Asyncio, ScriptsData structures, classes, exception handling, multithreading, and modular architecture across real production scripts.
Speech Recognition & TTS
ElevenLabs & pyttsx3Real-time microphone capture, hotword detection, Google Speech API binding, and natural neural voice playback.
Basic PyTorch
Tensors & Local InferenceNeural models inference, tensor operations, CUDA device targeting, and local MusicGen loading via Hugging Face.
OpenCV
Computer VisionReal-time video processing, facial detection cascades, color thresholding, and morphological image transformations.
FastAPI
High-Speed MicroservicesAsynchronous REST routing, Pydantic schema validation, automatic OpenAPI / Swagger documentation, and dependency injection.
API Implementation
REST, JSON, WebhooksIntegration with Groq, ElevenLabs, Cloudflare Workers AI, Magic Hour, and external weather/news APIs.
Basic SQL
MySQL & Relational DesignRelational table creation, CRUD queries, indexing, and persistent database connections for web booking systems.
Dev Tools & Git
Linux, VS Code, GitHubVersion control branching, PyAutoGUI desktop automation, virtual environments, and `.env` secret management.
Developer Terminal
Got a project?
Let's build it together.
Looking for a dedicated Python backend developer for voice AI, API integrations, PyTorch models, or MySQL database pipelines? Drop me a line anytime!