> whoami
Suprovo Mallick
> class
AI / ML Developer
> specialization
GenAI systems • RAG pipelines • agentic workflows • ML deployment
> mission
Build useful, deployable, and visually cool AI systems instead of dead notebook demos.
> current_focus
- retrieval engineering
- grounded generation
- runtime verification
- FastAPI + LLM backends
- scalable AI product workflows
> location
West Bengal, India
| Signal | Value |
|---|---|
| 🎓 Education | B.Tech CSE, KIIT (2023–2027) |
| 💼 Current Direction | GenAI apps, RAG, agents, production AI |
| ⚡ Strength | Turning ideas into deployable systems |
| 🧠 Problem Solving | 280+ DSA problems solved |
| 🏅 Recognition | HackerRank 5-Star Python |
| 🎯 Vibe | Hacker energy + AI builder mindset |
system: AI wheat disease detection
stack: [TensorFlow, EfficientNetV2, FastAPI, Docker]
accuracy: 93%+
status: deployed
LIVE NODE: https://wheat-analysis-app.vercel.app |
system: AI diet + wellness companion
stack: [LangChain, Gemini API, ChromaDB, FastAPI]
type: RAG assistant
status: live
LIVE NODE: https://aahar-react.vercel.app |
system: multi-agent architecture generator
stack: [LangGraph, Groq LLaMA, Streamlit]
mode: orchestration
status: deployed
LIVE NODE: https://intent-compiler-bydyno.streamlit.app SOURCE: https://github.com/DYNOSuprovo/intent-compiler |
system: multilingual translation engine
stack: [Transformers, PyTorch, FastAPI]
model: NLLB-200
latency_gain: 38%
LIVE NODE: https://huggingface.co/spaces/Dyno1307/Translate-V2 |
+ Add dedicated public repos for Wheat Guardian, Aahar, and Pluto
+ Then this section can become a full 4-card showcase grid
flowchart LR
A[User Query] --> B[Semantic Retrieval]
B --> C[Vector / Context Search]
C --> D[Agentic Reasoning]
D --> E[Grounding + Verification]
E --> F[FastAPI / UI Delivery]
flowchart TD
A[Problem Idea] --> B[Design]
B --> C[Prototype]
C --> D[Model Integration]
D --> E[Backend API]
E --> F[Deployment]
F --> G[Iteration]
+ Experimenting with retrieval optimization
+ Building RAG-backed assistants
+ Designing agentic system flows
+ Improving grounding and verification
+ Shipping AI through APIs and frontendslab_focus:
- semantic retrieval
- vector search
- LLM orchestration
- FastAPI deployment
- inference optimization
- production-minded AI systems
[system_profile]
alias = DYNO
mode = cyberpunk_ai_builder
favorite_stack = Python + FastAPI + GenAI
energy_source = curiosity + problem solving + cool demos
preferred_output = systems_that_actually_work
+ I don't build dead demos.
+ I build AI systems with movement.
+ I build pipelines that deploy.
+ I build products that feel alive.
+ I build cool things because boring is illegal.