01 / Robotics & autonomyHackathon prototype
SUTRA
Exploring how a coordinated drone swarm could support search and rescue when navigation and communication become unreliable.
ROS 2 / PX4 / C++ / Python
A work in curiosityHyderabad, India
Hello, I’m Nikhil.
Curiosity,made tangible.
A nerd exploring the intersection of
robotics, AI, and a better future.
The current coordinates
Robotics & physical AIAI-augmented developmentLearning from first principles
Read my résumé Robotics / Intelligence / Software Scroll to discoverSelected explorations, 2025—26
01 / The possibility
I believe the most interesting future happens where intelligence meets the physical world.
So I build, experiment, break things, and try again.
These projects are a few steps along that path.
02 / Selected workBuilt, explored, documented.
The work.
A closer look at the systems I build —
and the thinking behind them.
02 / On-device AI & mobile softwarePreparing for release
Vitt
An on-device AI finance assistant connecting local language-model inference, transaction organization, and a private financial ledger.
Flutter / Dart / LiteRT / Gemma
03 / Local AI & developer toolsPublished on PyPI
Saara CLI
A published local-first CLI and SDK for researching topics, preparing datasets, labeling examples, and validating training data.
Python / Firecrawl / Ollama / vLLM
04 / Computer vision & edge AIPublished on Hugging Face
AerialEye
A published YOLOv11-nano model for aerial and disaster-response imagery, paired with tiled inference for small-object detection.
Python / PyTorch / YOLOv11 / SAHI
03 / The person behind the projects
Still learning. Always building.A curious mind,
A curious mind,
learning by building.
I’m Kilani Sai Nikhil, a second-year B.Tech Computer Science student at NxtWave Institute of Advanced Technology, based in Hyderabad. I’m drawn to the moment an idea stops being abstract and starts doing something in the world.
I combine hands-on Python, SQL, Linux, and data work with AI-assisted development. MCP tools, custom agent skills, and AI pair programming help me explore faster; first-principles practice and testing help me understand what I’m building.
A little more about meAI-augmented development
Build with AI.
Think for myself.
- My hands-on foundation
- Python · SQL · Pandas · NumPy · GCP · Docker · Linux · Git
- My AI development workflow
- MCP servers · custom agent skills · tool connectors · AI pair programming · multi-agent workflows
- Frameworks I build with AI assistance
- React / Next.js · Flutter / Dart · PyTorch · YOLO · ONNX · TFLite
- What I’m strengthening independently
- Python DSA · linear algebra · backpropagation · runtime debugging · agent tool contracts
04 / The direction2025 → 2029
Two ways forward
Deep technical foundations and practical building, growing together.
Track A / The foundationsStart with
Start with
first principles.
Python & SQL · linear algebra · DSA
Neural networks · systems & debugging
Track B / The buildingBuild with
Build with
AI alongside.
MCP tools · custom agent skills
AI pair programming · local model workflows
Where the tracks meet
Robotics × AI × useful software
Make ideas work. Understand why they work.
Now / In progress
Connecting the disciplines
Local AI tools, on-device applications, and robotics projects. Strengthening Python algorithms, linear algebra, agent tool contracts, and independent debugging.
A direction I’m working toward. Future research, releases, and internships are goals.
On the research horizon
NeuroNav-ROS2
A planned research capstone exploring neuromodulated local planning: connecting differential-drive kinematics, learned policies, and ROS 2 simulation. Currently at the foundations and planning stage.
05 / Learning & recognitionThe complete record
Always a student.
Courses, certifications, and projects that keep the learning moving.
01 / 12 — Introduction to Neural Networks
Scroll through the complete record.
06 / Start a conversation
Have a question, an idea, or a wonderfully hard problem?