A practical toolkit for taking an AI idea from interface to infrastructure.
Languages used across AI pipelines, web applications, and data-processing projects.
What I use them for
Applied projects in visual document understanding, computer vision, and time-series forecasting.
What I use them for
Building LLM-backed workflows with structured outputs, prompt iteration, evaluation, and practical guardrails.
What I use them for
Designing APIs, services, auth flows, and data pipelines that support real product behavior.
What I use them for
Building usable product interfaces with typed React stacks, responsive layouts, and AI feature surfaces.
What I use them for
Working with retrieval, embeddings, vector storage, model evaluation, and practical data workflows.
What I use them for
Deploying and maintaining projects with CI/CD, managed hosting, containers, and cloud-provider services.
What I use them for
Designing voice-agent flows with practical conversation paths, validation, and fallback behavior.
What I use them for
Practical exposure to web security testing and threat modeling, with introductory smart contract assessment experience.
What I use them for
Here's what those tools look like in practice.
Ingest documents, build a retrieval path, pass useful context to the model, then evaluate the result.
See AndhaKaanunBring the interface, APIs, auth, data, and AI provider together in one usable flow.
See Email Digital TwinCompare prompts, check structured outputs, and catch regressions before they become product behavior.
See LLM Evaluation Framework