Skip to content
View KripaMishra's full-sized avatar
🎯
Focusing
🎯
Focusing
2 followers · 4 following

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair Extraordinairex3Achievement: Pull Sharkx3Achievement: YOLOAchievement: Quickdraw

Block or report KripaMishra

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
KripaMishra/README.md

hey, I'm Kripa 👋

I build AI things that are useful after the demo.

Agent systems · healthcare AI · local-first developer tools · too many terminal tabs

LinkedIn Email Profile views

whoami

I'm an AI engineer from India, building healthcare AI at Tap Health. I turn fuzzy product ideas into systems people can actually trust: clear behavior, useful context, safe fallbacks, and zero hand-wavy magic.

Outside health workflows, I'm usually building agent tooling, testing retrieval systems, or making another CLI for a problem that probably deserved a script.

values: useful AI > flashy demos · boring reliability > clever fragility · shipping > slide decks

toolbox

Languages & apps

Python TypeScript FastAPI Node.js React

AI, search & agents

OpenAI LangChain RAG MCP Hugging%20Face Milvus Elasticsearch SQLite

Systems

PostgreSQL MongoDB Kafka Docker Kubernetes AWS

things I've been making

Project What it does Built with
Scavenger Routes one-shot AI tasks to capable free-tier providers. Less spend, more useful answers. Python · CLI · LLMs
Research Mode Systematic research with scoped investigation, evidence ledgers, and adjustable depth. Python · Agents · Research
CUDA Documentation RAG Citation-first copilot for CUDA documentation — answers with receipts. Python · RAG · Milvus · BM25
Career Copilot Job discovery, resume tailoring, and browser-assisted applications. TypeScript · AI agents
DocNeedle Deterministic local-first doc lookup, stale-term checks, and corpus evaluation. Python · SQLite FTS5
pi-acp An ACP adapter for the pi coding agent. TypeScript · ACP

contribution telemetry

Kripa's GitHub contribution and streak stats
Kripa's contribution graph

If you're building trustworthy AI, better developer tools, or healthcare products that don't feel like paperwork — let's talk.

Pinned Loading

  1. research-mode research-mode Public

    Pi skill for systematic, repeatable online research with: grill-first scoping Firecrawl-first discovery/fetching fallback web fetch/search when Firecrawl fails or lacks coverage adaptive Quick / St…

    Python 1

  2. scavenger scavenger Public

    A lightweight Python CLI that routes one-shot AI tasks to free-tier providers, maximising utility without spending money.

    Python 1

  3. cuda-documentation-rag cuda-documentation-rag Public

    CUDA Documentation Copilot is a citation-first Retrieval-Augmented Generation (RAG) demo for CUDA documentation questions.

    Python

  4. career-copilot career-copilot Public

    Personal career assistant for job discovery, resume personalization, and browser-assisted applications.

    TypeScript

  5. doc-needle doc-needle Public

    Local-first deterministic doc lookup/update prototype using SQLite FTS5. DocNeedle has two separate modes of use: Normal CLI usage: index a markdown tree, search it, read a chunk, run stale-term ch…

    Python 1

  6. mastra-pii mastra-pii Public

    Layered PII redaction for Mastra agents — Presidio container or local deterministic engine, Indian ID recognizers (Aadhaar/PAN/UPI/IFSC) with checksums, fail-closed fallback. No prompt changes.

    TypeScript