Available for new opportunities

Portfolio — 2026

Siddharth Patni

|

Agentic AI & Full-Stack Engineer pursuing M.Sc. in Digital Technologies. Building autonomous LLM pipelines, observability platforms, and cloud-connected systems with Python, FastAPI, and Playwright.

View My Work Get In Touch
Python FastAPI LangGraph Playwright OpenAI Anthropic PyTorch Docker Kubernetes PostgreSQL Redis Next.js TypeScript React Hugging Face OpenRouter

AI & Autonomous Systems

LLMs, RAG Pipelines, Computer Vision, LangChain — Designing and deploying intelligent autonomous systems at production scale

LLM + Vision + Autonomous AI

MLOps & Model Deployment

MLflow, DVC, Docker, Kubernetes — Building reproducible ML pipelines and deploying models with CI/CD automation

Training + Deployment + Monitoring

Full-Stack Development

React, FastAPI, Node.js, PostgreSQL — Building production-grade web applications from frontend to backend

Frontend + Backend + Database

DevOps & Cloud Infrastructure

Docker, Kubernetes, Jenkins, GitHub Actions — Orchestrating CI/CD pipelines and cloud infrastructure

AWS + Azure + GCP
01 — Profile

About Me

I'm an AI & Autonomous Systems Engineer focused on building production-grade intelligent systems — from LLM-powered applications and RAG pipelines to computer vision systems for real-world autonomous decision-making.

Currently pursuing my M.Sc. in Digital Technologies at TU Clausthal & Ostfalia University, I combine deep ML knowledge with applied engineering. My projects span AI-driven safety systems, intelligent chatbots, and automated content pipelines — each built with proper MLOps practices, containerization, and CI/CD.

I don't just train models — I architect end-to-end AI solutions, from data ingestion to production deployment, with a strong focus on reproducibility, scalability, and real-world impact.

1+
Year Professional Experience
10+
GitHub Projects
15+
Technologies
3
Cloud Platforms

Currently Based In

Braunschweig, Germany

Lower Saxony

Languages

English C1 (Advanced)
German A2-B1 (In Progress)
02 — Education

Education

Mar 2025 - Present

M.Sc. in Digital Technologies

TU Clausthal & Ostfalia University

Germany

Specializing in Information & Communication Technologies with focus on MLOps, AI infrastructure, and advanced automation systems.

Current
Jul 2021 - May 2024

B.Tech in Computer Engineering

Charotar University of Science and Technology

India

CGPA 7.6
Jun 2018 - Jun 2021

Diploma in Computer Engineering

Gujarat Technological University

India

High Distinction
CGPA 8.92
03 — Experience

Research & Work Experience

SoSe 2026 · Ongoing
Research · Ostfalia / TU Clausthal

Agentic AI for Automated Web Data Extraction

CORE Research Group — Industry Cooperation

Braunschweig, Germany

  • Built LLM-based scraper generation pipelines with Python, Playwright and multi-model LLM APIs (OpenRouter) for scalable extraction
  • Implemented self-healing feedback loops with sandboxed code execution to verify script reliability across runs
  • Benchmarked cost, latency and success-rate trade-offs across LLM providers to identify production-ready models
  • Explored Computer-Use Agent (CUA) architectures and contributed to a cascaded FastAPI pipeline with async job processing and persistent document storage
Python Playwright FastAPI LLM APIs OpenRouter Agentic AI
Dec 2023 - Apr 2024

Software Developer Intern

L&T Technology Services

India

  • Dockerized services and integrated them into Linux-based CI pipelines, reducing regressions and improving release consistency
  • Reworked React components and Flask APIs, improving perceived load times by ~30%
  • Harmonized API-UI contracts and added validation checks to reduce integration defects
Docker React Flask CI/CD
Jun 2023 - Jul 2023

Data Analyst Intern

Snapfix Ltd.

India

  • Built Python/SQL ETL pipelines to standardize datasets, reducing report preparation time by ~40%
  • Authored comprehensive handover documentation (data dictionaries, pipeline steps) to ensure reproducibility
Python SQL ETL Pandas
04 — Selected Work

Featured Projects

Selected work across agentic AI, LLM infrastructure and full-stack systems.

Featured

Sentinel

LLM Observability · Open Source

Drop-in proxy infrastructure for OpenAI and Anthropic models with full execution traces, evaluation layers and smart routing — built for production LLM workloads.

  • Comprehensive execution traces & evaluation layers
  • Smart routing across OpenAI / Anthropic providers
  • Compliance-ready EU AI Act audit logs & real-time alerts
Python FastAPI OpenAI Anthropic Observability
Featured

CereForge

AI Learning Platform

Adaptive learning platform powered by LLM-generated content pipelines. Personalizes study paths based on user progress and identified knowledge gaps.

  • LLM-generated adaptive content pipelines
  • Async backend with streaming responses & task queues
  • Automated evaluation of open-ended AI solutions
Python LLMs AsyncIO Docker Redis

Workflow Engine

Visual No-Code Builder

Node-based visual workflow editor for building multi-step automations with production-scale support for conditional branching and REST API integrations.

  • Drag-and-drop node editor for complex flows
  • Conditional branching & REST API integrations
  • Designed for production-scale execution
TypeScript React Node.js PostgreSQL

ShopMate-R

AI Retail Assistant

AI-powered shopping assistant with natural-language product search. Integrates LLM reasoning with structured product data for Pepper and Temi service robots.

  • Natural-language product search & recommendations
  • LLM reasoning over structured catalog data
  • Deployed on Pepper & Temi robotics platforms
Python LLMs Robotics NLP

AI Video Production Pipeline

AI + Automation · 2024

Automated captioning, media retrieval and GPU-accelerated rendering pipeline. Reduced production turnaround from 6 hours down to 20 minutes.

  • FastAPI orchestration with Redis-backed queues
  • Hugging Face models for captioning & retrieval
  • GPU-accelerated FFmpeg rendering pipeline
Python FastAPI OpenCV Hugging Face FFmpeg Redis

AI Driver Safety System

Computer Vision

Deep learning system with real-time traffic sign recognition, pedestrian detection and driver monitoring for road safety.

YOLO CNN OpenCV TensorFlow

Multi-Disease Diagnostic App

Medical AI · 2023

End-to-end diagnostic pipeline with ~85% accuracy; reduced per-case processing time from 30 minutes to under 10 seconds.

Python Flask OpenCV scikit-learn

LLM Chatbot

AI · Full-Stack

Modern chatbot with Flask, React and HuggingFace Transformers. Real-time message streaming and context-aware responses.

Flask React HuggingFace
05 — Toolbox

Technical Skills

Interactive Skills Proficiency Map

Core AI & MLOps

MLflow DVC Model Cards LangChain RAG Pipelines Hugging Face Vector DBs OpenCV TensorFlow PyTorch

DevOps & Infrastructure

Docker Kubernetes Helm Charts CI/CD GitHub Actions Terraform Linux Prometheus Grafana

Programming & Data

Python Bash SQL Pandas FastAPI REST APIs JSON/YAML

Documentation & Operations

Technical Runbooks Service Inventories Incident Management Confluence Notion AI Governance
06 — Contact

Let's Connect

Get In Touch

I'm always open to discussing new opportunities, collaborations, or just chatting about AI, DevOps, and technology.

Location Braunschweig, Germany

My Documents

Resume

Professional resume — AI & Autonomous Systems Engineer

Cover Letter

General cover letter for AI & software engineering roles