Engineering ideas into intelligent systems.
AI/ML engineer and SIH 2025 winner passionate about building practical AI projects, exploring emerging technologies, and solving meaningful problems through engineering.
About Me
I am an AI/ML engineer with a strong interest in artificial intelligence, machine learning, deep learning, and software development. I enjoy transforming complex ideas into practical and meaningful solutions.
Through hands-on projects and the Smart India Hackathon, I have developed skills in Python, AI-powered applications, and collaborative problem-solving. I am passionate about exploring emerging technologies, building innovative systems, and continuously growing as an AI engineer.
Systems & Platforms
A selection of production-shaped machine learning solutions, infrastructure pipelines, and deep learning implementations.
EVALUATION APPARATUS & PROMPT VERSION LINEAGE
Chronicle
AI-powered Multi-LLM platform for prompt versioning, A/B testing, and execution tracking.
- Built a PostgreSQL-backed prompt version control system with cryptographically hashed, immutable version histories and append-only release promotion audits.
- Engineered a database-first execution pipeline logging exact token counts, latency (86ms P50), and fractional API costs before dispatching remote calls, ensuring crash-safe auditability.
- Implemented a multi-objective evaluation engine utilizing Pareto Frontier analysis and the geometric knee-point heuristic to automatically recommend the optimal cost-vs-accuracy model configuration.
BOTANICAL PATHOLOGY & CELLULAR LESION DETECTION
LeafLens
Multilingual AI agricultural diagnostics platform detecting 41 distinct crop diseases.
- Engineered a convolutional neural network (CNN) classifier achieving 98.4% test accuracy across 41 crop-disease categories.
- Developed a localized mobile interface supporting 11 regional languages, ensuring accessibility for non-technical farmers.
- Shipped real-time edge inference on mobile devices with local caching, bypassing connectivity bottlenecks in remote fields.
VISUAL SPEECH RECOGNITION & PHONETIC WAVE ARCS
LipSense
Deep learning visual speech recognition model translating lip movements to text.
- Trained a spatio-temporal CNN coupled with Recurrent GRU layers, obtaining a 95.2% sentence-level transcription accuracy.
- Reduced word error rate (WER) to 4.8% on target grid vocabulary benchmarks through custom visual landmark pre-processing.
- Built an optimized pipeline extracting lips ROI at 30fps using lightweight face mesh trackers.
BIOMETRIC FACE LIVENESS & VECTOR EMBEDDING SEARCH
Facelogix
High-performance facial authentication and security platform utilizing advanced facial embedding distance.
- Built a secure local authentication gateway using FaceNet/InsightFace embeddings, reaching 99.2% verification accuracy.
- Implemented a anti-spoofing mechanism utilizing texture analysis and eye-blink frequency detection to block static image bypasses.
- Optimized facial verification latency to sub-200ms using Redis vector caches for local employee indices.
SEMANTIC TEXT INTELLIGENCE & TRANSFORMER NODES
OmniText
Advanced NLP dashboard and text analysis toolkit running local transformer models.
- Integrated lightweight BERT and RoBERTa models locally for zero-cost sentiment analysis, named entity recognition (NER), and text classification.
- Created a highly efficient batch-processing pipeline for text summarization running on CPU instances.
- Designed a clean editorial dashboard demonstrating multi-document comparisons and semantic overlap metrics.
SimAstra: Autonomous Swarm Defense Framework
Ministry of Defence / Bharat Electronics Limited (CRL) · Dec 2025
SWARM DEFENSE & DECENTRALIZED FLIGHT TRAJECTORIES
As a member of Team Palantier, I contributed to the development of SimAstra, an autonomous swarm-defense framework created for Smart India Hackathon 2025 under Problem Statement SIH25164.
The project explored decentralized coordination for autonomous drone networks, enabling swarm agents to share information, coordinate defensive actions, and adapt to dynamic operational conditions.
CONTRIBUTIONS & OUTCOMES
- Collaborated with Team Palantier to secure the National Winner title under Problem Statement SIH25164.
- Contributed to the development and simulation of decentralized swarm-coordination and autonomous task-allocation approaches.
How I Learn & Build
I approach projects by understanding the problem, exploring the technology behind it, and turning ideas into practical solutions. I enjoy learning through experimentation, research, and hands-on development.
From AI and machine learning to intelligent systems and emerging technologies, I focus on building projects that strengthen my technical skills while solving meaningful problems.
Problem-First Thinking
I begin by understanding the problem, the users, and the purpose of the solution before selecting technologies or designing the implementation.
Learn by Building
I explore new concepts through research, experimentation, and hands-on projects, turning technical knowledge into practical experience.
Iterate & Improve
I treat every project as a learning process—testing ideas, identifying limitations, improving the implementation, and building better solutions over time.
Education & Involvement
Academic credentials and community service milestones at the University of Mumbai.
B.E. in Computer Science (Artificial Intelligence & Machine Learning)
Nov 2022 – Jun 2026Theem College of Engineering, University of Mumbai
- CGPA: 8.22/10
- Focus areas: Artificial Intelligence, Machine Learning, Neural Networks, Natural Language Processing, and Software Architecture
NSS Volunteer / Member
2024 – 2025Theem College of Engineering NSS Unit
- Organized community development programs, health camps, and environmental awareness drives
- Worked in team environments to execute government-sponsored social welfare projects
Capabilities & Credentials
Structured catalog of engineering skills, model families, cloud systems, and professional industry certifications.
Skills Inventory
Languages & Frameworks
GenAI & LLMs
Data & Cloud Platforms
Certifications
AWS Data Engineering
Amazon Web Services (AWS)
Disaster Risk Monitoring with AI
NVIDIA
AI Fundamentals & AI-Enabled Applications
IBM
Google Cloud Skills Arcade
Google Cloud
Green Skills & AI
Edunet / Microsoft
Intro to ML & Feature Engineering
Kaggle
Let's discuss systems & models.
I am actively looking for AI Engineer, ML Engineer, or Python developer positions starting mid-2026. If you have an open role or want to discuss my work, drop me an email.