Skip to content
View nk-droid's full-sized avatar
🎯
Focusing
🎯
Focusing
25 followers · 68 following

Achievements

Achievement: QuickdrawAchievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLO

Achievements

Achievement: QuickdrawAchievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: YOLO

Block or report nk-droid

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.

Content in all repositories owned by your account will be closed.
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
nk-droid/README.md

Nidhish Kumar

AI Systems Engineer

Building production LLM workflows, RAG evaluation systems, agentic automation, and backend AI infrastructure.

Portfolio LinkedIn Hashnode

What I build

I build AI systems that move beyond prototypes and survive real users.

Currently working as SDE – AI/ML at 2Sigma School, where I build and integrate LLM-powered pipelines for grading, content generation, adaptive learning, evaluation, and backend automation.

Recent work:

  • Built and integrated 11 production AI pipelines for an adaptive learning platform
  • Designed an automated grading pipeline that reduced manual grading effort by ~70%
  • Built self-healing generation pipelines that reduced malformed-output failures by ~94%
  • Built systems across LLM workflows, RAG, evaluation, backend automation, and observability

Core stack

AI / LLM: LangChain, LangGraph, RAGAS, Transformers, OpenAI SDK, Google GenAI SDK, Ollama
Backend / Infra: Python, FastAPI, Pydantic, SQLAlchemy, PostgreSQL, Redis, Ray, Docker
Retrieval / Evaluation: FAISS, BM25, pgvector, Chroma, OpenTelemetry, Prometheus, Grafana, Tempo
Cloud / Frontend: GCP Cloud Run, Cloud Storage, Cloud SQL, React, MUI

Writing

I write about building AI systems with evaluation, reliability, and production constraints in mind.

Read my blogs on Hashnode →

Selected achievements

  • GATE 2024 DA — All India Rank 379
  • Winner — AI-Agent Hackathon, IIT Madras
  • Finalist — Smart India Hackathon, ISRO Challenge
  • Top 5 Finalist — Dr. Reddy’s Digital Health Hackathon, among 1900+ teams

Pinned repos show my current flagship work: AutoPR and Modular RAG.

Pinned Loading

  1. AutoPR AutoPR Public

    Production-style autonomous coding agent that converts GitHub issues into tested PRs using LangGraph, Ray, Redis queues, PostgreSQL, observability, and HITL governance.

    Python

  2. ModelMesh ModelMesh Public

    A Kubernetes-native AI model gateway that intelligently routes requests across multiple LLM providers with autoscaling, load balancing, and unified OpenAI-compatible APIs.

    Python 1

  3. rag rag Public

    Config-driven RAG evaluation workbench for composing, running, and comparing retrieval pipelines across context quality, faithfulness, latency, and answer quality.

    Python

  4. ScanPlus ScanPlus Public

    JavaScript