Introduce

I am Brijesh,
Data Scientist and Travel Photographer

From Automotive Engineer to Data Scientist — I turn ambitious AI visions into production reality. Specializing in LLMs, agentic systems, and end-to-end ML solutions that drive measurable business impact.

6+

years of
Experience

15+

Projects &
Hackathons

6+

Blogs

⚡ FYI: I vibe coded this entire website in under 6 hours with GitHub Copilot using a single prompt.

About Me

From automotive engineer to AI leader — my journey embodies the power of strategic career pivoting and continuous learning. Starting as a Quality Assurance Manager at Tata Motors, I recognized the transformative potential of AI and made a bold decision to pursue a Master's in Germany, completely reimagining my career trajectory.

Today, I'm a seasoned Data Scientist with 6+ years specializing in LLMs, agentic AI systems, and production ML deployments. At Alexander Thamm GmbH, I led developments and cross-functional teams, architecting ML pipelines that drove measurable business impact for Fortune 500 clients. Now at Hilti, I continue pushing the boundaries of AI innovation in the construction domain.

What sets me apart: I don't just build models — I deliver end-to-end AI solutions that solve real business problems. From proof-of-concepts to production systems, I've guided 15+ projects through their complete lifecycle. My international perspective, having worked across India and Germany, brings unique insights to global AI challenges.

Beyond the code: I'm passionate about knowledge sharing through my AI blog, mentoring emerging data scientists, and capturing the world through travel photography — because creativity fuels innovation. Whether you're looking for fractional AI leadership, need an experienced consultant for your next LLM project, or seeking a strategic hire who can scale your AI initiatives, I bring both technical depth and business acumen to every challenge.

Education & Experience

Experience

Nov 2024 - Present
Data Scientist
Hilti, Kaufering
May 2022 - Oct 2024
Principal Data Scientist & Reinforcement Learning Lead
Alexander Thamm GmbH, Munich
Jun 2020 - Apr 2022
Algorithm Developer
ZF Mobility Solutions GmbH, Ingolstadt
Mar 2019 - Feb 2020
Reinforcement Learning Expert
ZF Mobility Solutions GmbH, Ingolstadt
Aug 2015 - Aug 2017
Vehicle Quality Assurance Manager
Tata Motors Limited, Ahmedabad

Education

2018 - 2020
M.Engg. Automotive Engineering
Technische Hochschule Ingolstadt
2011 - 2015
B.Tech. Mechanical Engineering
National Institute of Technology Surat

Tools

VS Code
VS Code
GitHub Copilot
GitHub Copilot
Claude Code
Claude Code
Ollama
Ollama
Python
Python
JavaScript
JavaScript
HTML
HTML
CSS
CSS
SQL
SQL
Kotlin
Kotlin
AWS
AWS
Azure
Azure
Databricks
Databricks
Terraform
Terraform
Docker
Docker
GitHub
GitHub
GitLab
GitLab
GitHub Actions
GitHub Actions
PyTorch
PyTorch
TensorFlow
TensorFlow
Scikit-learn
Scikit-learn
Hugging Face
Hugging Face
OpenAI
OpenAI
OpenAI Gym
OpenAI Gym
Flask
Flask
FastAPI
FastAPI
Streamlit
Streamlit
Gradio
Gradio
LangChain
LangChain
CrewAI
CrewAI
Jupyter
Jupyter
Plotly
Plotly
Weights & Biases
Weights & Biases

GitHub Repositories

Smartypy
Build your own local Python code assistant with Llama-3 and Ollama.
Terraform AWS Website Deployer
One-click Terraform pipeline to host static sites on AWS for ~€20/year.
Agentic Blogger
Multi-agent LLM framework to generate full blog posts from a topic prompt.
Getting Started with GANs (MNIST)
Keras/TensorFlow notebooks to learn and demo GANs on handwritten digits.

Blogs

How to build your personal coding assistant in 10 mins
Spin up a llama3-based Python code assistant with Ollama and connect it to VS Code in minutes.
Read Article
Coding your first GAN algorithm with Keras
A step-by-step guide to building a GAN from scratch in Keras/TensorFlow and training it on MNIST.
Read Article
Introduction to Generative Adversarial Networks (GANs)
A simple intro to GANs, their generator–discriminator game, and real-world uses like image synthesis.
Read Article
Reinforcement Learning – Framework and Application Example
A comprehensive guide bridging basic RL understanding with practical problem-solving using Q-Learning and the deadly maze example.
Read Article
Reinforcement Learning – Deadly Triad
Deep dive into the critical concept of deadly triad in RL: function approximation, bootstrapping, and off-policy learning challenges.
Read Article
Prompt Optimization with Reinforcement Learning in LLMs
Exploring how RL can optimize prompts in large language models through trial-and-error learning for better AI outputs.
Read Article

Contact