Open to opportunities

kabir mehta

@kabirmehta

Message

Applied AI Engineer building reliable production AI agents and LLM applications with RAG and orchestration.

Germany

Message

What I'm looking for

I’m looking to build TypeScript-based, multi-agent LLM systems with production vector search and strong monitoring. I’d rather work on reliable agentic workflows than just “working once,” and a remote Germany setup suits me well.

I build AI agents and LLM applications that run in production, using multi-agent orchestration with LangGraph and RAG systems over live data sources. I care about reliability, not just demos—what I monitor and catch before it reaches users.

At HeidelAI, I shipped agentic features end to end: multi-agent orchestration, LLM application delivery, and FastAPI backends on AWS with Docker. I also improved production robustness by finding a non-deterministic retrieval issue via Langfuse traces, then fixing it with deterministic seed pinning and updated monitoring rules.

I’m comfortable in both Python and TypeScript, and my DevMind project (AI Code Intelligence with Vector Search and Multi-Agent Architecture) is the one I’m most proud of technically. I hold an M.Sc. in Data Science and Analytics from SRH Hochschule Heidelberg, and I bring a disciplined, engineering-first approach to verifying AI outputs.

Experience

Work history, roles, and key accomplishments

Current

Machine Learning Engineer

Current

HeidelAI

Jan 2024 - Present (2 years 6 months)

Built end-to-end agentic AI features including LangGraph orchestration and RAG pipelines over live data, served via FastAPI backends on AWS. Implemented monitoring with Langfuse to improve reliability and caught non-deterministic retrieval behavior before it reached users.

LangGraph RAG fastAPI AWS Docker Langfuse Vector Search Embeddings Python TypeScript GitHub Actions CI CD Monitoring Deterministic Seed Pinning PostgreSQL

Data Science Consultant (Case Study)

ONTEC AG

Jan 2023 - Present (3 years 6 months)

Built reproducible machine learning pipelines on large-scale industrial datasets and delivered structured outputs and insights to business stakeholders using clear evaluation criteria.

Python Machine Learning Machine Learning Pipelines Model Evaluation

Education

Degrees, certifications, and relevant coursework

SRH Hochschule Heidelberg

Master of Science (M.Sc.), Data Science and Analytics

2022 - 2024

Completed an M.Sc. in Data Science and Analytics at SRH Hochschule Heidelberg (2022–2024), translating product requirements into shipped AI/LLM software and debugging model behavior in production-like environments.

Kurukshetra University

Bachelor of Technology (B.Tech), Computer Science

2018 - 2022

Earned a B.Tech in Computer Science from Kurukshetra University (2018–2022), with an emphasis on validating and verifying AI outputs rather than assuming correctness.