Skip to content

AI × Human·Chennai, Tamil Nadu, India

Nagasubramaniam G

AI/ML Engineer | GenAI and LLM/RAG Systems | Full-Stack Developer

Building Intelligent Systems. Engineering Real-World Solutions.

I build practical AI-powered applications and full-stack products, combining machine learning, generative AI, data and software engineering to solve real-world problems.

Nagasubramaniam G — professional portrait
Nagasubramaniam G — professional portrait

Nagasubramaniam G

About

Human story behind the systems

I am an early-career AI/ML engineer who builds practical GenAI systems and production-ready web products — from problem framing through architecture, implementation, and deployment.

I completed a B.Tech in Computer Science with an Artificial Intelligence and Machine Learning specialisation at Kalasalingam Academy of Research and Education (August 2021 – May 2025). Alongside formal study, I trained at Bosscoder Academy in data science and machine learning.

My work sits at the intersection of applied AI and product engineering: retrieval-augmented assistants, computer-vision experiments, solo SaaS architecture, and freelance production websites. I care about clear ownership, verified evidence, and systems that ship.

I am looking for AI/ML, GenAI, LLM/RAG, and applied AI engineering roles — and carefully scoped freelance engagements where I can own delivery end to end.

01 · 2021 – 2025

B.Tech CSE — AI & ML

Kalasalingam Academy of Research and Education. Specialisation in Artificial Intelligence and Machine Learning.

02 · Training

Bosscoder Academy

Data Science and Machine Learning programme.

03 · Practice

Freelance & solo product work

Live client sites for Chola Power and Dharani Fitness, SportsHub SaaS architecture, and collaborative GenAI projects.

04 · Recognition

Hack-AI-thon · Patent · Publication

First place among 39+ teams; Indian Patent No. 573184; fourth-author research at ICMCTT 2024.

Skills

Interactive capability map

Capabilities grouped by domain. Labels reflect project evidence where available — not invented proficiency scores.

Constellation

AI and Machine Learning

Core ML and deep-learning capabilities backed by project work where noted.

  • Machine LearningWorking knowledge
  • Deep LearningWorking knowledge
  • PyTorchProject-backed
  • Scikit-learnWorking knowledge
  • NLPWorking knowledge
  • Computer VisionWorking knowledge
  • Model EvaluationWorking knowledge

Services

What I can build for you

For founders and operators who need a production website, storefront, or practical GenAI feature — with clear ownership and live evidence where available.

Production websites

Fast, credible marketing and business sites with modern Next.js delivery, responsive UI, and deploy hygiene (CDN, forms, SEO basics).

  • Public launch-ready sites
  • Mobile-first layouts
  • Technical SEO foundations

E-commerce storefronts

Catalogue, cart, checkout, payments, and account flows for product businesses — scoped to what the engagement actually includes.

  • Product discovery UX
  • Checkout / payments integration
  • Admin-friendly catalogue paths

GenAI & LLM/RAG apps

Practical conversational and retrieval systems: embeddings, vector search, prompt/evaluation loops, and FastAPI or Next.js service layers.

  • Grounded Q&A assistants
  • Retrieval pipelines
  • Evaluation-minded iteration

Full-stack product builds

End-to-end web products that combine UI, APIs, auth, data models, and deployment — including AI-assisted features where they earn their place.

  • Auth and role-aware flows
  • Postgres-backed data models
  • Docker / cloud deployment

Live client delivery

Public evidence: cholapower.in · dharanifitness.in

Start a project

Selected Work

Evidence over claims

Flagship case studies lead — including live freelance delivery. Additional team and research projects follow with written case studies. Solo, freelance, and team ownership stay explicit.

Concept visual of a sports facility booking dashboard with venues, calendar, and time slots.

01 · Solo

SportsHub

Solo Product | Full-Stack SaaS

SportsHub is a production-oriented sports facility management platform designed to bring discovery, booking, access control and payment workflows into one coherent product. I designed and delivered the system end to end — from the data model and role-based access controls to the customer experience, payment flow, and deployment pipeline. The platform uses more than 29 relational database models to support the operational complexity behind a real multi-role SaaS product.

Next.js · TypeScript · React · PostgreSQL · Prisma · Clerk v6

Open case study
Desktop screenshot of the live Chola Power homepage. Brand and design belong to Chola Power Private Limited.

02 · FreelanceLive

Chola Power

Chola Power is a live digital platform for a solar-energy business. I delivered a fast, trustworthy web experience that helps a real company communicate its services, build credibility, and support customer enquiries — pairing a polished front end with deployment, deliverability, security, and discoverability concerns that are often missing from portfolio-only work.

cholapower.in

Case study
Concept visual of a document assistant with retrieval diagram and grounded conversation.

03 · Team

Conversational AI Assistant with RAG

A conversational AI assistant that combines large language models with semantic retrieval to produce more grounded, useful responses. The system was built around Llama and Mistral models, FAISS-powered retrieval, Sentence Transformers embeddings, and a FastAPI service layer. Iterative prompt and retrieval tuning produced an F1 score above 89 percent.

Case study
Desktop screenshot of the live Dharani Fitness homepage.

04 · FreelanceLive

Dharani Fitness

Freelance e-commerce website for a gym-equipment shop. The public site covers product discovery, with cart, checkout, Razorpay, Google sign-in, customer accounts, order history, and product administration included in the engagement.

dharanifitness.in

Case study

Additional work

Supporting evidence with visuals — team classifications stay explicit.

Team · Recognised

Multimodal AI Virtual Assistant

An award-winning multimodal generative AI assistant built during a 24-hour hackathon, unifying voice, visual, and text workflows into a functional real-time assistant that won 1st place among 39+ competing teams.

Team · Recognised

Skin Cancer Prediction and Detection

A computer-vision decision-support system leveraging transfer learning and image preprocessing to assist in early skin-lesion assessment — framed as clinical assistance, not autonomous diagnosis.

Team · Recognised

Flower Plucking Robotic Arm

A patented cross-disciplinary mechatronics and computer-vision system engineered to detect mature flowers and execute precision automated harvesting.

Team

Amazon Web Scraping and Product Intelligence

An end-to-end data engineering and product analytics pipeline that extracts publicly visible e-commerce listings and transforms them into structured, analysis-ready market intelligence.

Team

Air Quality Monitoring System

An IoT environmental sensing and visualization platform that captures real-time air-quality sensor signals and translates raw telemetry into an accessible, human-readable monitoring dashboard.

Team

Java Library Management System

A foundational object-oriented Java application built to streamline core library operations — catalog indexing, member records, borrowing lifecycles, and real-time availability tracking.

Team

Accident Risk Prediction

A supervised machine-learning workflow that models historical road-incident data to estimate accident risk levels across environmental, temporal, and roadway conditions.

Team

IoT Carbon Sequestration Monitor

A sustainability-focused IoT telemetry system built to monitor and log chemical carbon-sequestration performance in a sodium hydroxide (NaOH) capture setup.

Team

YOLO Transportation Analytics

A real-world computer-vision analytics pipeline utilizing YOLO object detection on custom traffic datasets to identify, classify, and analyze vehicles in complex road scenes.

Team

Emotion Aware Music Recommendation

A multimodal affective computing experiment that captures facial emotion cues and maps user sentiment to personalized music recommendations — assistive personalisation, not clinical emotion diagnosis.

Team

Roadside Object Safety Detection

A safety-focused object detection system designed to spot roadside hazards and near-vehicle obstacles that drivers frequently overlook — scoped as assistive perception research, not certified ADAS.

Team

Continuous Sign Language Translation

An inclusive-AI video understanding pipeline that translates continuous sign-language gestures into coherent, usable text output.

Team

Predictive Machine Failure Detection

A predictive-maintenance analytics system that monitors multivariate industrial sensor time series to flag abnormal operating patterns prior to equipment breakdown.

Experience

Delivery without invented job titles

No formal employment history is recorded. This section reflects freelance delivery, solo product work, collaborative AI projects, education, and training.

freelance

Client engagement

Production website delivery

Chola Power · Freelance

  • Delivered a live Next.js 15 / React 19 website with Docker, Cloudflare CDN, DDoS protection, transactional email, and technical SEO.
  • Focused on trustworthy public presence for a solar-energy business.

freelance

Client engagement

E-commerce full-stack delivery

Dharani Fitness · Freelance

  • Built a public storefront with catalogue, cart, checkout, Razorpay, Google sign-in, accounts, and product administration.
  • Live at dharanifitness.in.

product

Solo product

Solo SaaS product engineering

SportsHub

  • Owned architecture through deployment for a multi-role facility booking platform.
  • 29+ relational models, Clerk RBAC, Razorpay, and AI slot recommendations.

collaboration

Collaborative project

LLM / RAG systems

Conversational AI Assistant · Team

  • Contributed to retrieval, embeddings, model interaction, evaluation, and FastAPI integration.
  • Reported F1 above 89% after iterative tuning.

education

August 2021 – May 2025

B.Tech Computer Science — AI & ML

Kalasalingam Academy of Research and Education

  • Specialisation in Artificial Intelligence and Machine Learning.
  • No formal employment history is recorded — experience is project and freelance based.

training

Technical training

Data Science and Machine Learning

Bosscoder Academy

  • Structured training in data science and machine learning alongside degree work.

Credentials

Recognition, research, and certifications

Verified awards, patent attribution, publication authorship, Microsoft/NASSCOM/Oracle credentials, and NPTEL recognition — only claims backed by certificate records.

Award

Hack-AI-thon — First Place

KARE ACM & Diebold Nixdorf Hack-AI-thon (Dec 2024). 24-hour competition, 1st place among 39+ teams, ₹5,000 prize. Multimodal AI Virtual Assistant — team contribution.

Patent

Indian Patent No. 573184

Flower Plucking Robotic Arm. Appl. 202241055456 (Sept 2022), granted Oct 2025 under Kalasalingam Academy of Research and Education. Attribution matches the official patent record. FlowerBot research-paper authorship is not claimed.

Publication

Earlier Cancer and Tumor Detection Using Deep Learning Techniques

ICMCTT 2024. Fourth author. Official proceedings details follow the verified publication record.

Certificate proof

Scans of key Microsoft and NPTEL appreciation certificates. Course PDFs remain on file for verification.

NPTEL Motivated Learner certificate of appreciation for Nagasubramaniam G, JAN–APR 2025.
NPTEL Motivated Learner
NPTEL Enthusiast certificate of appreciation for Nagasubramaniam G, JUL–DEC 2024.
NPTEL Enthusiast
Microsoft Certified Azure AI Fundamentals certificate for Nagasubramaniam G.
Azure AI Fundamentals
Microsoft Certified Azure Fundamentals certificate for Nagasubramaniam G.
Azure Fundamentals

Recognition

NPTEL Motivated Learner — JAN–APR 2025

IIT Madras / SWAYAM Certificate of Appreciation. Criteria: appeared in ≥8 exams (Jul 2023–Jan 2025), passed ≥75% of courses appeared, and passed at least one exam in each of four semesters.

Recognition

NPTEL Enthusiast — JUL–DEC 2024

IIT Madras / SWAYAM Certificate of Appreciation. Criteria: appeared in ≥8 exams (Jan 2023–Jul 2024) and passed ≥75% of courses appeared.

Certification

NASSCOM AI — Data Scientist: Visualizations V3.0

NSQF Level 6. Certificate No. CN-0949070. Score 93.47. Issued via Bosscoder Software Services Private Limited.

Certification

Microsoft Certified: Azure AI Fundamentals (AI-900)

Earned 5 Mar 2023. Credential ID ECC83FCCCC9B7020 · Certification No. 391BBH-B0144B. Online verifiable Microsoft credential.

Certification

Microsoft Certified: Azure Fundamentals (AZ-900)

Earned 12 Mar 2023. Credential ID 18ECEB333CC03FE2 · Certification No. BEN526-3BD0CB. Online verifiable Microsoft credential.

Certification

Time-bounded

Oracle Cloud Data Management 2023 Certified Foundations Associate

Earned 5 Jul 2024. Credential 100711164OCDMF2023. Valid until 5 Jul 2026 — recorded as time-bounded, not presented as permanently current.

Coursework

Selected NPTEL / SWAYAM certifications

Verified course certificates include Soft Skills, Soft Skill Development, Project Management, Leadership and Team Effectiveness, Forests and their Management, Conservation Economics, Conservation Geography, Wildlife Ecology, and One Health.

Skills

HackerRank Python and SQL

Python and SQL practice credentials on HackerRank. Public badge URLs can be linked when profile visibility is confirmed.

Contact

Let’s talk about a role or project

Hiring managers: share the role and stack. Founders and clients: share the website, storefront, or GenAI scope. I usually reply within two business days.

Open to AI/ML roles and scoped freelance web / GenAI projects.

Prefer the form for structured enquiries, or reach out directly by email, phone, or WhatsApp.