[ About ]
PRANEETH
I study AI systems, write rigorous notes, and build tools that make research easier to inspect.
Scroll for the signal ↓/ Now
NUS Computer Science + Mathematics. AI Society Tech and Research Director.
I spend most of my time around interpretability, model evaluation, agent reliability, and the engineering layer that turns research taste into working software.
Read my research taste →/ Method
Read deeply. Build small. Make the mechanism visible.
My default loop is simple: study a paper or system, extract the mechanism, build a minimal implementation, and write down the tradeoffs clearly enough that someone else can challenge them.
Browse the notes →/ Proof
Public work beats long biographies.
Instead of front-loading every credential, this site points outward: selected projects, essays, repositories, and research trails carry the detail.
/ Selected work
A few systems worth inspecting first.
Static knowledge system / JavaScript
Notes
Great, no-nonsense computer science field notes
Agentic workflow / ShellAgentCoding
A public Shell repository in Praneeth Suresh's GitHub portfolio.
Applied software tool / ShellOpenAIBuild
A public Shell repository in Praneeth Suresh's GitHub portfolio.
AI and ML research / Jupyter NotebookLOBForecasting
A public Jupyter Notebook repository in Praneeth Suresh's GitHub portfolio.
Public repository / UnspecifiedPraneeth-Suresh
Config files for my GitHub profile.
Public repository / TypeScriptspill
A public TypeScript repository in Praneeth Suresh's GitHub portfolio.
/ Conversations
Useful conversations start from concrete technical overlap.
Reach out about AI engineering, applied ML, developer tools, ML systems, interpretability, efficient inference, agent reliability, internships, or NUS AI Society collaboration.
Email me →/ Repository map
Project terrain
AI and ML research
Software and app systems
Energy, data, and experiments
Curiosity is only useful when it becomes a system someone else can understand, run, and build on.
/ Contact
Contact me about research, internships, consulting, or NUS AI Society collaboration.
Useful messages usually start from a specific overlap: a paper, project, team, startup problem, internship role, workshop, sponsor idea, or AI Society collaboration.
Email is the best first step. Include the context, timeline, and the kind of response that would be useful.