Looking for a job

Rizeen Musadhik

@rizeenmusadhik

ML Research Engineer at SEVEN PRINCIPLES GROUP building GraphRAG and multi-agent tender workflows that cut screening time by 58%.

Germany

Message

What I'm looking for

I’m interested in research that moves from architecture and training to measurable system performance, especially in multimodal and video understanding, temporal reasoning, parameter-efficient adaptation, retrieval and agentic systems, and reliable GenAI for regulated domains.

At SEVEN PRINCIPLES GROUP, I built a GraphRAG pipeline and multi-agent system to automate German public-sector tender assessment. The workflow reduced manual screening time by 58% and enabled the team to submit 30% more relevant bids.

For my M.Sc. thesis at University of Hildesheim, I extended Qwen2.5-VL-7B with cross-frame attention for temporal video understanding and developed a distributed, two-stage training pipeline. I also designed evaluation probes to distinguish motion understanding from benchmark shortcuts.

At sovanta AG, I built citation retrieval for a production RAG product and developed an SAP BTP solution-architect agent with prompt optimisation and validation. At Medulla – Veguar Technologies, I co-founded a medical community startup and increased user engagement by 42% with a hybrid recommender.

Experience

Work history, roles, and key accomplishments

Current

Werkstudent Applied AI Engineer

Current

SEVEN PRINCIPLES GROUP

Feb 2026 - Present (8 months)

Sole AI engineer automating end-to-end assessment of German public-sector tenders, replacing manual document review with an automated bid/no-bid decision pipeline. Built a GraphRAG document-intelligence pipeline and a multi-agent bid-assembly system, reducing manual screening time by 58% and increasing relevant bid throughput by 30%.

Graph RAG Multi Agent Systems MCP MongoDB Python fastAPI PDF Processing Microsoft GraphRAG

Werkstudent Data Science

sovanta AG

Feb 2025 - Aug 2025 (6 months)

Built citation retrieval for a production RAG product and developed an SAP BTP Solution-Architect agent. Built an LLM scraping and validation pipeline and implemented GenAI observability for FastAPI/LlamaIndex pipelines.

LlamaIndex SAP UI5 Selenium Pydantic OpenTelemetry ArizePhoenix fastAPI RAG

Co-Founder & Data Scientist

Medulla – Veguar Technologies

Aug 2021 - Feb 2023 (1 year 6 months)

Co-founded a medical community startup as lead data scientist and represented the company in investor pitches. Increased user engagement by 42% with a hybrid recommender and reduced churn by 36% with an ensemble model.

BERT LDA Matrix Factorization Random Forest Gradient Boosting SMOTE SHAP A B Testing

M.Tech Thesis

IIT Kharagpur

Jan 2020 - Jan 2021 (1 year)

Developed an LSTM achieving 87.7% on raw data from 128 sensor channels, matching classical models built on 561 hand-engineered features.

LSTM Sensor Data Representation Learning

Student Research Project

GitHub

Developed Boosted Fully Connected Networks by adapting BoostNet from vision to tabular data and integrated them into the TabSurvey benchmark. Benchmarked against XGBoost, LightGBM, CatBoost, and deep-tabular baselines on five datasets.

BoostNet TabSurvey XGBoost LightGBM CatBoost Optuna Python

M.Sc. Thesis

University of Hildesheim

Extended Qwen2.5-VL-7B with a cross-frame attention block in all 32 vision-encoder layers for temporal understanding. Designed a two-stage distributed DeepSpeed pipeline and evaluated on VITATECS, TempCompass, and MVBench.