Kilani Sai Nikhil
Computer Science student at NxtWave (Class of 2029) engineering local-first on-device AI architectures (Vitt), dataset distillation engines (saara-ai), quantized aerial vision models (AerialEye), and biometric data pipelines (Sleep Health Analytics).
Engineering Identity & Profile
Engineering at the Intersection of Edge ML & Systems Architecture
I am a Computer Science student at NxtWave (Class of 2029) based in Hyderabad, India. My engineering philosophy combines rock-solid native CS fundamentals (C++, Python, SQL, Linux, Docker, PyTorch) with 10x AI-augmented velocity (MCP orchestration, LiteRT on-device inference, semantic code graphs).
I focus on building software that runs directly on edge hardware—eliminating recurring cloud subscription costs, safeguarding personal privacy through zero-cloud architectures, and maintaining resilience during severe infrastructure disruption.
Kilani Sai Nikhil // Profile Overview
I am pursuing a Bachelor of Technology in Computer Science & Engineering at NxtWave Institute (2025–2029). I engineer local-first software systems, open-source AI tooling, and edge computer vision models.
- Native Systems Mastery: Deep proficiency in Python, C++, SQL relational modeling, Linux OS internals, and Docker containerization.
- 10x AI Augmentation: Accelerated scaffolding, semantic code-review graph traversal, and multi-agent MCP toolchain orchestration.
- Edge Privacy Stance: 100% on-device inference via LiteRT & Google Coral TPUs with zero external cloud dependencies.
Core Engineering Disciplines
On-Device AI & LiteRT
Deploying quantized small language models (Gemma 4 E2B) locally on mobile hardware with zero cloud latency, zero recurring API cost, and strict privacy.
Edge Vision & Disaster CV
Fine-tuning YOLOv11 with SAHI dynamic tile slicing for small high-altitude objects, quantized to INT8 precision for low-power Google Coral TPUs.
Data Science & Biometrics
Conducting statistical exploratory data analysis, feature engineering, and distribution modeling on clinical datasets using Pandas, NumPy, and Seaborn.
Open-Source Tooling & MCP
Publishing production developer SDKs and dataset distillation engines on PyPI and NPM with clean architecture and automated CI/CD pipelines.
Engineering Systems & Deployments
Vitt (Artha)
100% Private, On-Device AI Financial Assistant
100% private on-device AI financial assistant for MSMEs running Gemma 4 E2B via LiteRT with zero SMS permissions.
- On-device local LLM inference via LiteRT (Gemma 4 E2B) at ~14 tokens/sec with zero cloud bills
- Zero SMS or contact permissions: scrapes banking alerts locally via Android Notification Listener
saara-ai
Local-First Dataset Generation & Distillation Engine
Local-first dataset generation CLI on PyPI & NPM with 2,600+ downloads, 38 releases, and multi-agent web scraping.
- Dual published on PyPI and NPM with 2,600+ downloads and 38 stable releases
- Interactive terminal CLI wizard (`npx saara-ai wizard`) for zero-configuration dataset synthesis
AerialEye
YOLOv11-Nano Disaster Response & Aerial CV Model
YOLOv11-Nano aerial object detection model with SAHI slicing, 89.4% mAP@0.5, and INT8 Coral TPU quantization.
- Published on Hugging Face Hub with 1,900+ downloads across weights and dataset
- Curated unified dataset of 6,327 high-altitude aerial and disaster-response images
Sleep Health Biometrics & Analytics
Statistical Biometric Modeling & Lifestyle Sleep Health Analysis
Statistical biometric EDA and sleep health analysis modeling correlations between stress, sleep quality, and cardiovascular indicators.
- Comprehensive biometric data pipeline cleaning, feature engineering, and blood pressure decomposition
- Statistical correlation analysis identifying key lifestyle drivers behind sleep quality and stress levels
Dual Capability Matrix
Core Languages & Systems
Hand-engineering core algorithms, memory management, file descriptor operations, multi-threading, and system utilities on Linux.
Rapid scaffolding of interface definitions, type-safe API boundaries, and runtime interop bridges using AI agent loops.
Databases & Relational Modeling
Designing strict relational schemas, complex joins, indexing strategies, and hardware-encrypted local databases compliant with DPDP Act 2023.
Automating database migrations, synthesizing test seed fixtures, and setting up hybrid relational/vector search layers.
Edge ML & Computer Vision
Curating aerial vision datasets, architecting SAHI sliding-window inference, and quantizing weights to INT8 for sub-watt micro-TPU execution.
Orchestrating system-level on-device language models via LiteRT and packaging automated CI/CD model card pipelines.
Developer Tooling & Environments
Deploying self-hosted container stacks, configuring local documentation scrapers, and structuring robust CI/CD release pipelines.
Designing autonomous multi-agent delegation loops, custom MCP servers, and dynamic prompt distillation workflows (SFT / DPO).
Data Science & Statistical Modeling
Cleaning multi-dimensional biometric datasets, decomposing compound metrics, imputing missing values, and engineering structured pipelines.
Generating multi-variable distribution pairplots, correlation matrices, and automated statistical report synthesis.
Core Languages & Systems
Hand-engineering core algorithms, memory management, file descriptor operations, multi-threading, and system utilities on Linux.
Rapid scaffolding of interface definitions, type-safe API boundaries, and runtime interop bridges using AI agent loops.
Databases & Relational Modeling
Designing strict relational schemas, complex joins, indexing strategies, and hardware-encrypted local databases compliant with DPDP Act 2023.
Automating database migrations, synthesizing test seed fixtures, and setting up hybrid relational/vector search layers.
Edge ML & Computer Vision
Curating aerial vision datasets, architecting SAHI sliding-window inference, and quantizing weights to INT8 for sub-watt micro-TPU execution.
Orchestrating system-level on-device language models via LiteRT and packaging automated CI/CD model card pipelines.
Developer Tooling & Environments
Deploying self-hosted container stacks, configuring local documentation scrapers, and structuring robust CI/CD release pipelines.
Designing autonomous multi-agent delegation loops, custom MCP servers, and dynamic prompt distillation workflows (SFT / DPO).
Data Science & Statistical Modeling
Cleaning multi-dimensional biometric datasets, decomposing compound metrics, imputing missing values, and engineering structured pipelines.
Generating multi-variable distribution pairplots, correlation matrices, and automated statistical report synthesis.
Verified Credentials & Honours
Google Cloud Gen AI Master Certificate
Master verification demonstrating enterprise proficiency across Vertex AI, Generative AI models, multimodal agent deployment, and RAG architectures.
Build & Deploy Agents in Production
Advanced engineering badge covering production-grade agent deployment patterns, monitoring, observability, and scaling on Google Cloud.
Agents 101 Certification
Foundational certification covering AI agent architectures, tool-use patterns, cognitive loops, and multi-agent coordination concepts.
Deploy Your First Agent
Hands-on validation badge for deploying autonomous AI agents to Google Cloud infrastructure using Vertex AI tooling.
Intermediate SQL
Advanced relational queries covering grouping with HAVING, multi-table joins, subqueries, set operations, and relational query tuning.
Introduction to SQL
Structured Query Language fundamentals — SELECT queries, JOINs, filtering, aggregations, and relational database normalization.
National Project Showcase Finalist (Vitt)
Selected as a national buildathon finalist for engineering Vitt: 100% on-device AI financial tracking compliant with DPDP Act 2023.
Contact
Let's build something real.
I am always open to collaborate on open-source packages, on-device machine learning architectures, and edge systems engineering. Drop me a line directly or connect on registries.