Let’s Build Together
Clear communication, fast iterations, and a bias for shipping. You always know where the project stands.
ML · Agentic AI · Automation Engineer
DATASCIENTIST
Final-year Big Data student at SRM University AP, already shipping production AI/ML — real-time computer vision, forecasting models, and GenAI agents that run end-to-end with zero manual steps.
- Projects Shipped
- 15+
- Hackathons
- 8+
- Internships
- 3
Projects Shipped
Hackathons
Internships
Machine Learning, Forecasting & Computer Vision
Agentic AI · RAG · LangGraph · Automation
How I work
WHAT YOUGET
What You Get
Clean, documented code, models with real metrics, and deployed solutions — not notebooks that die on my laptop.
End-to-End Ownership
From data scraping and model training to APIs, dashboards, and automation pipelines that run themselves.
Live in the wild · Computer Vision
Real-time AI,actually shipped.
Rakshak turns raw CCTV into a live stampede-risk signal — detecting hundreds of people per frame and forecasting danger 30 minutes out. Not a notebook demo; a system built to run.
- Concurrent users in production
- 500+
- People tracked per frame
- 200–500+
- Forward risk forecasting
- 30 min
Concurrent users in production
People tracked per frame
Forward risk forecasting
Selected work
RECENTPROJECTS
01Computer Vision· Dec 2025 – Jan 2026
Rakshak
Real-time crowd monitoring & stampede prevention
A real-time crowd-safety system that detects dangerous density build-up and forecasts stampede risk before it happens.
- Python
- YOLOv8
- BoT-SORT
- OpenCV
- FastAPI
- MongoDB
- WebSocket
- 200–500+
- People / frame
- 7+
- Risk metrics
- 30 min
- Forecast window
02Machine Learning· Mar – Apr 2026
AgileAI
AI-augmented agile project management
An ML-driven MERN platform that predicts sprint-failure risk, auto-estimates task effort, and flags developer burnout before it crosses critical thresholds.
- Python
- XGBoost
- FastAPI
- Scikit-learn
- Optuna
- React
- Node.js
- F1 0.92 · 95.2%
- Sprint Risk
- MAE 0.8 · R² 0.89
- Effort Est.
- F1 0.93 · 97.1%
- Burnout
03GenAI · Automation· Feb – Mar 2026
Riya — AI Property Broker
Multi-channel RAG assistant that books site visits end-to-end
A full-stack, multi-channel AI property assistant that guides buyers from first interest to a confirmed site visit over WhatsApp, Telegram, or web (Live Demo: https://144-24-156-187.sslip.io) — while keeping the broker in the loop.
- Python
- Claude API
- Groq
- LangChain
- pgvector
- PostgreSQL
- WhatsApp Cloud API
- WhatsApp · Web · TG
- Channels
- pgvector RAG
- Retrieval
- Oracle Cloud VM
- Deploy
04Machine Learning· Sep – Oct 2025
InvenIQ
Demand forecasting & inventory intelligence for quick-commerce
An end-to-end demand-forecasting and inventory system for Blinkit-style Indian dark stores — LightGBM forecasting wired into a multi-agent LangGraph pipeline and a real-time dashboard.
- Python
- LightGBM
- Optuna
- Pandas
- Scikit-learn
- LangGraph
- Next.js
- 0.88
- R² (variance)
- 91.4%
- Stockout Acc.
- 22%
- Cost cut
05GenAI · SaaS· 2026
Multi-Tenant WhatsApp Agent
Cloud-native AI support & sales bots as a service
A SaaS platform where many companies run isolated AI-powered WhatsApp support and sales bots from a single deployment (Live Demo: https://multi-tenant-whatsapp-agent.vercel.app) — each with its own persona, catalog, and knowledge base.
- FastAPI
- LangGraph
- Groq Llama 3.3 70B
- Gemini 2.0 Flash
- ChromaDB
- MongoDB Atlas
- React
06GenAI · Automation· 2026
InvoiceFlow
Self-correcting AI email automation for Gmail
An AI email automation system that reads inbound Gmail, categorizes it, drafts grounded replies, proofreads itself, and either sends automatically or queues sensitive mail for human review.
- LangGraph
- Gemini 2.0 Flash
- ChromaDB
- HuggingFace Embeddings
- FastAPI
- React 19
- TypeScript
Where I've shipped
WORKEXPERIENCE
Seqtoo Softwares
Jul – Dec 2025 · RemoteSoftware Intern (PHP Developer)
Thajaa — E-Commerce Backend
Rebuilt an entire grocery-platform backend from scratch: RESTful APIs, order lifecycle & tracking, RBAC, inventory, cart logic, and payment gateway on PHP/MySQL.
500+ concurrent users in production
EduHub — Government School Platform
Led end-to-end development of a production multi-tenant SaaS for Telangana government schools: RBAC dashboards, CMS, digital notice board, admissions, and timetables.
Deployed to 5+ schools · sub-100ms
EduBot
2025 · RemoteAI Engineer Intern
EduBot — AI Learning Assistant
Built a conversational AI learning assistant that answers student questions from course material, using retrieval-augmented generation over an LLM with a chat interface.
Teachnook
Jun – Jul 2024 · RemoteWeb Development Intern
EduConnect — Peer Mentorship Platform
Learners had no structured way to discover mentors or schedule peer sessions. Built end-to-end: user profiles, skill listings, rule-based mentor matching, and calendar booking system with real-time search, dynamic filtering, and 5+ integrated frontend modules across auth, discovery, and scheduling flows.
The toolkit
TECHSTACK
Languages
Programming- Python
- Java
- JavaScript
- TypeScript
- PHP
- C++
- SQL
Data & Analytics
Analysis- Pandas
- NumPy
- EDA
- Power BI
- Matplotlib
- Seaborn
- Jupyter
Machine Learning
Modelling- Scikit-learn
- XGBoost
- LightGBM
- Feature Engineering
- Time-Series
GenAI & LLMs
Agents & RAG- Claude API
- OpenAI
- LangChain
- LangGraph
- RAG
- FAISS
- ChromaDB
- CrewAI
- Groq
Computer Vision
Perception- YOLOv8
- OpenCV
- ByteTrack
- BoT-SORT
Backend & DB
Services- FastAPI
- REST
- MySQL
- PostgreSQL
- MongoDB
- Snowflake
Automation
Pipelines- Airflow
- n8n
- GitHub Actions
- Google APIs
- Streamlit
DevOps
Ship- Docker
- Git
- GitHub
- Linux / Shell
Receipts
CERTIFICATIONS& CREDENTIALS
A running dump of everything I've earned — Oracle Cloud AI, the Vodafone Idea Foundation ML program, IBM SkillsBuild, DataCamp, and hands-on internships. Verify the badges or open the certificate itself where I've got it on file.
VOIS · Vodafone Idea Foundation
AI / ML Program- Introduction to Machine LearningVOIS × Connecting Dreams · 2025
- Performance Evaluation in Machine LearningVOIS · 2025
- Exploratory Data Analysis for MLVOIS · 2025
- Machine Learning ModelsVOIS · 2025
- Neural Networks with PythonVOIS · 2025View
- Building Neural Networks with KerasVOIS · 2025View
- Introduction to Computer VisionVOIS · 2025View
- Sentiment Analysis with ML & NLPVOIS · 2025View
- Machine Learning for Text MiningVOIS · 2025View
- ML in Biomedical Signal ProcessingVOIS · 2025View
- Data Collection using SensorVOIS · 2025View
- Artificial Intelligence in PracticeVOIS · 2025View
- AI Design ChallengesVOIS · 2025View
Hackathons
CodeStrom AIBest UI AwardPixelSaladRunner-UpByteVerse 7.0NIT PatnaTesserXNIT PatnaTechXcelerateHack With GujaratAditya Birla Engenuity2025
From the desk
DESIGNTHOUGHTS
- Machine Learning
How gradient-boosted trees actually work
XGBoost feels like cheating until the day it does something weird. Here is the from-scratch mental model — residuals, similarity scores, gain, and why the learning rate is really a humility knob.
Jun 14, 2026 · 10 min read
- Machine Learning
Graph-based anomaly detection for time series
Real systems are never one wiggly line. When dozens of series move as a system, the interesting failures are broken relationships — and that is exactly what a graph of time series is built to see.
Jun 2, 2026 · 11 min read
- Machine Learning
XGBoost vs LightGBM: which gradient booster should you reach for?
They both win Kaggle competitions, but they grow trees in fundamentally different ways. A practical, from-experience breakdown of when leaf-wise beats level-wise — and when it quietly overfits you.
May 18, 2026 · 9 min read
- Computer Vision
Building Rakshak: real-time crowd safety with YOLOv8
Counting 500 people in a frame is the easy part. Turning those boxes into a trustworthy NORMAL / WARNING / CRITICAL signal — before a stampede — is where the real engineering lives.
Feb 2, 2026 · 11 min read
- GenAI
A zero-hallucination RAG broker on a 100% free stack
How I wired BeautifulSoup, FAISS, and a small LLM into a property assistant that never invents listings — and then let it book its own site visits through the Google Calendar and Gmail APIs.
Mar 20, 2026 · 10 min read
LET’S WORKTOGETHER
Open to opportunities · usually replies within a day
Open to Data Science, ML, Agentic AI, and automation roles — internships, freelance builds, or a good problem worth solving. Fastest way to reach me is email, or grab a 30-minute Zoom.