Hello, I'm

Somitra Singh
Kushwah.

Applied AI & backend engineer building agentic LLM systems that actually ship to real users.

M.S. CS, UT Arlington (May 2026) · Open to full-time roles Arlington, TX
See my work Get in touch

01. About

I just graduated from UT Arlington with an M.S. in Computer Science (May 2026), and I'm an engineer who's spent the last couple of years building things people actually use.

Lately, most of my energy goes into agentic AI — LangGraph workflows, MCP tools, multi-agent orchestration, and the messy operational reality of running these systems on a VPS with real traffic. I'm interested in the part where models stop being demos and start being products: tool design, evaluation, prompt iteration, failure modes.

Before grad school I shipped backend and data systems at Red Hill Labs — Node, DynamoDB, MERN, AWS. That foundation is why I can take an LLM idea from notebook to a hardened endpoint with logs and restarts that survive a Tuesday morning.

I'm a tinkerer at heart. I self-host. I read papers. I rewrite my own tools instead of buying them. If something looks like magic, I want to take it apart.

02. Selected projects

OpenClaw & Hermes Agent

May 2026

Self-hosted agentic deployment · Linux VPS

Deployed production agent instances on a VPS, exposed through Telegram and Discord bots, and used them as daily drivers for code, research, and content workflows.

  • Configured MCP server integrations, tool permissions, and sandboxed execution.
  • Built async webhook + long-poll bridge in Python with per-chat session state and rate limiting.
  • Hardened with systemd, log rotation, reverse proxy, and secrets management — codified real failure-mode fixes into restart-safe configs.
PythonMCPTelegramDiscordsystemdNginx
Private repository

AI Hotel Booking Agent

Mar 2026

LangGraph multi-node agent · GPT-4o

A conversational booking agent built on LangGraph's stateful graph with conditional edge routing — a hands-on take on the ReAct pattern.

  • 4 custom LangChain tools with Pydantic-validated schemas, backed by parameterized SQL against PostgreSQL (Neon).
  • Persistent conversation memory via LangGraph's PostgresSaver, plus per-user long-term memory for preferences.
  • Streamlit chat UI with Google OAuth 2.0 (PKCE) and user-scoped data isolation.
LangGraphLangChainGPT-4oPostgreSQLStreamlitOAuth
View on GitHub

Mobile Price Prediction

Nov 2025

End-to-end ML pipeline · AWS SageMaker

A classification model predicting phone price ranges, with the full train/deploy/invoke loop running on SageMaker — S3 for data, IAM for least-privilege access.

  • Custom training script + SageMaker estimator for cloud training and evaluation.
  • Scalable prediction endpoint with a clean upload → train → deploy → invoke workflow.
SageMakerScikit-LearnS3IAMPython
View on GitHub

Network Security & Phishing Detection

Nov 2025

Flask web app · Dockerized · GitLab CI/CD

A Python network-security tool that ingests real-world datasets to detect phishing and flag suspicious patterns, wrapped in a Flask UI for analyst review.

  • Backend ingestion + validation scripts to flag anomalies for follow-up.
  • Containerized with Docker and shipped via GitLab CI/CD to reduce release friction.
FlaskDockerGitLab CIPython
View on GitHub

03. Experience

Software Engineer — Data & Backend Systems

Jan 2023 – Jul 2024

Red Hill Labs · Intern → Full-Time

  • Designed a centralized inventory + sales platform on Node.js and DynamoDB enabling real-time analytics over large-scale transactional data.
  • Built ingestion and reporting pipelines that improved data accuracy ~15% and cut report generation from hours to seconds.
  • Shipped MERN apps integrated with Elastic Beanstalk, DynamoDB, and S3 — owned production rollout and post-deploy hardening.
  • Wrote backend APIs with strict validation and idempotency, reducing data inconsistencies by ~30%.

M.S. Computer Science Graduated

Aug 2024 – May 2026

University of Texas at Arlington

B.Tech Computer Science

2019 – 2023

Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore

04. Toolbox

LLM & Agentic AI

LangChain, LangGraph, MCP, OpenAI & Anthropic APIs, tool/function calling, RAG, multi-agent orchestration, ReAct, Pydantic structured outputs, eval & observability.

Languages

Python, JavaScript, TypeScript, Java, C++, C, SQL.

Backend & APIs

FastAPI, Flask, Node.js, Express, REST, OAuth 2.0, PostgreSQL, SQLite, DynamoDB, MongoDB.

Cloud & Infra

AWS (EC2, S3, DynamoDB, Beanstalk, SageMaker, Bedrock), GCP basics, Oracle Cloud, Linux VPS, Nginx, systemd.

ML & MLOps

Scikit-Learn, PyTorch, HuggingFace Transformers, SageMaker, model deployment, MLOps pipelines.

DevOps & Tools

Docker, GitHub Actions, GitLab CI/CD, Git, Claude Code, Cursor.

05. Let's talk

I'm actively looking for forward-deployed and applied-AI engineering roles. If you're building something interesting at the messy intersection of LLMs and real-world systems — I'd love to hear about it.

somitrasinghkushwah@gmail.com +1 (682) 344-2942
GitHub LinkedIn