Computer Vision + NLP
Herbica AI
An AI assistant for medicinal plant identification that explains properties, safety notes, and usage guidance from an image-first workflow.
95% target accuracy | 200+ species dataset
TensorFlowFastAPIReactOpenCVRAG
Open to Work · AI Engineer
Nishanth JayaramanAI Workspace
AI & Data Science Graduate | AI Engineer | Agentic AI Builder | Intelligent Product Developer
I design intelligent products that turn research-grade AI into fast, useful, human-centered systems.
7+
AI Projects
Across GenAI, CV, NLP, and agents
1
IEEE Paper
Published research in deep learning
450+
LeetCode
Problem solving practice
100+
GitHub
Annual contribution target
Identity
Nishanth Jayaraman
AI & Data Science Graduate | AI Engineer | Agentic AI Builder | Intelligent Product Developer
IEEE
Published Research
Available
Open to Work
ARIA.ASSISTANT
A portfolio you can interrogate.
Hybrid AI with local intent routing, agentic navigation, and Llama 3.1 powered reasoning.
Assistant Core
Hybrid AI architecture
01 / Local knowledge
Structured JSON powers instant portfolio answers — zero API calls, zero latency.
02 / Navigation actions
Commands like 'open projects' or 'go to contact' scroll instantly inside the OS.
03 / Groq AI
Llama 3.1 via Groq handles open-ended questions with fast inference.
04 / Graceful fallback
If Groq is unavailable, the local knowledge layer keeps ARIA fully online.
05 / Voice layer
Web Speech API captures voice input; SpeechSynthesis reads answers back aloud.
AI
ARIA Portfolio Copilot
Online
ARIA online. I'm Nishanth's AI copilot.
What I can do:
• Explain his IEEE research on plant identification
• Deep dive into AI projects (Herbica, Interview Agent, TN Kalvi Hub)
• Summarize skills, education, and experience
• Help you reach Nishanth via email or LinkedIn
• Answer questions in voice mode
Quick starts: Tell me about his research | Explain Herbica AI | What's his education? | How to contact?
SYSTEM.IDENTITY
Builder, researcher, product engineer.
Nishanth works at the intersection of applied AI, intelligent product engineering, and research, transforming advanced technologies like LLMs, multimodal AI, RAG systems, and agentic workflows into practical, scalable products that solve real-world problems.
01 / Mission
The mission is practical impact: AI for education, agriculture, accessibility, hiring, creativity, and developer productivity.
02 / Research Interests
Efficient LLM fine-tuning · Agentic AI systems · Multimodal reasoning · RAG evaluation · AI for social good · Edge AI deployment
03 / Builder Mindset
Ship fast, measure honestly, learn faster, and keep the human in the loop.
04 / Availability
Available for Software Development, AI products, research collaborations, and full-time AI/ML engineering roles.
SYSTEM.EDUCATION
Academic foundation.
Formal training in AI and Data Science combined with research, competitions, and real-world product engineering.
Undergraduate Degree
B.Tech in Artificial Intelligence & Data Science
Sri Sairam Engineering College, Chennai
Oct 2022 – May 2026CGPA: 8.76 / 10India
Key Highlights
Recognized as Best Academic Performer (4th Semester)
IEEE Research Paper Publication in Computer Vision & Deep Learning
Hosted national conference on AI & Emerging Technologies
Active contributor to department AI research and student tech events
Relevant Coursework
01Agentic AI Systems
02Software Development
03Machine Learning
04Deep Learning
05Natural Language Processing
06Computer Vision
07Data Structures & Algorithms
08Database Management
09Cloud Computing
10Big Data Analytics
Program Focus
Specialized in applying artificial intelligence and data science to real-world problems — from vision systems and NLP to agentic workflows and product engineering.
SYSTEM.CAPABILITIES
Skill matrix with product instincts.
A dashboard view of the AI, engineering, and research capabilities behind the portfolio OS.
Machine Learning
92%Supervised learningUnsupervised learningFeature engineeringModel evaluationMLOps basics
Deep Learning
88%CNNsRNNsTransformersDiffusion modelsTensorFlowPyTorch
Generative AI
94%LLMsPrompt systemsRAGFine-tuningStable DiffusionEvaluation
Agentic AI
86%Tool usePlanning loopsLangChainLangGraphMulti-agent workflows
Frontend Development
84%ReactNext.jsTypeScriptTailwind CSSFramer Motion
Backend Development
82%FastAPINode.jsREST APIsMongoDBPostgreSQLAuth flows
APIs and Deployment
80%VercelDockerGroqOpenRouterCloud APIsCI/CD
Research
78%IEEE writingExperiment designLiterature reviewTechnical documentation
SYSTEM.PROJECTS
Featured builds as explorable modules.
Each card opens like a product case file: problem, architecture, workflow, challenge log, outcomes, and future improvements.
Agentic AI
AI Interview Practice Partner
A conversational mock interview system that adapts question difficulty, scores answers, and generates improvement roadmaps.
500+ session target | 40% confidence lift
Next.jsGroqLangChainWeb Speech APIPostgreSQL
EdTech AI
TN Kalvi Hub
A learning hub for Tamil Nadu students with personalized study paths, bilingual explanations, and curriculum-aware AI support.
2,000+ student target | 85% outcome lift
ReactNode.jsMongoDBLLMsTamil NLP
NLP
Resume-JD Matcher
An ATS-style semantic matcher that compares resumes to job descriptions and explains gaps with improvement suggestions.
90% matching target | 60% review-time reduction
PythonBERTFastAPIPostgreSQLReact
SYSTEM.TIMELINE
Experience timeline.
Technology Job Simulation
Deloitte – Forage Platform
Developed a Python solution to normalize telemetry data from multiple JSON formats into a unified structured schema, simulating real-world data engineering tasks at enterprise scale.
Machine Learning Engineer Intern
Cloud IT Solutions
Developed a Generative Adversarial Network (GAN) model using TensorFlow to generate synthetic data, improving data augmentation efficiency by 20%. Designed an Image Modification Percentage Calculator in Python, improving image processing accuracy by 25% through statistical analysis and feature engineering.
SYSTEM.RESEARCH
Research, achievements, and signal.
Publications, certifications, awards, and evidence of consistent technical growth.
IEEE · 2024
Medicinal Plant Leaf Identification Using Transfer Learning and Deep Learning Techniques
This paper presents a transfer learning approach for automated medicinal plant leaf identification, combining deep convolutional neural networks with domain-specific feature extraction to achieve high-accuracy classification with limited training data. The system demonstrates practical deployment potential for healthcare, agriculture, and biodiversity research.
Transfer LearningDeep LearningComputer VisionPlant ClassificationMedicinal PlantsCNNs
Research Interests
Efficient LLM fine-tuningAgentic AI systemsMultimodal reasoningRAG evaluationAI for social goodEdge AI deployment
Achievements
Published an IEEE research paper on medicinal plant leaf identification using transfer learning and deep learning techniques
Built and deployed AI-powered applications including interview simulation agents and computer vision systems
Solved 450+ LeetCode problems with strong foundations in data structures and algorithmic problem solving
Hosted and coordinated a national conference on advancements in AI and emerging technologies
Organized large-scale technical and strategy-based events involving team management and budget planning
Recognized as Best Academic Performer during the 4th semester of B.Tech AI & DS
Winner of essay writing competitions and runner-up in intra-college public speaking competitions
Completed certifications and training programs from NPTEL, Coursera, Udemy, IIT Bombay Spoken Tutorial, and IIRS
SYSTEM.CODING
Competitive coding telemetry.
450+
LeetCode solved
DSA consistency and interview prep
100+
GitHub contributions
Public build habit
90 days
Coding streak
Daily problem-solving loop
7 builds
Project velocity
Shipped AI prototypes
SYSTEM.KNOWLEDGE
Tech stack and field notes.
A polished tool console paired with AI blogs, project writeups, research summaries, and learning notes.
Technology Orbit
PythonJavaSQLTensorFlowPyTorchScikit-learnPandasNumPyReactNext.jsTailwind CSSFramer MotionFastAPIFlaskNode.jsMongoDBLLMsAgentic AICNNsMachine LearningDeep LearningPower BITableauGitVercel
Learning Note
How I Think About RAG Evaluation
A practical checklist for measuring retrieval quality, answer faithfulness, and user trust.
Project Writeup
Building Herbica AI
Design notes on multimodal plant identification, safety constraints, and AI UX.
AI Blog
Agentic AI Without the Hype
A grounded view of tool use, planning, memory, and where agents become useful.
SYSTEM.CONNECT
Contact console.
Ready for AI products, research collaborations, and product engineering conversations.
Available · Open to Work
Build something intelligent, useful, and unusually polished.
Available for Software Development, AI products, research collaborations, and full-time AI/ML engineering roles.