Open to opportunities

Akshat Pandey

@akshatpandey1

Message

Research Associate advancing clinically interpretable computer vision and agentic LLM workflows for CT radiology.

India

Message

What I'm looking for

I’m looking for a research- and engineering-focused role in medical AI, where I can build clinically interpretable computer vision and agentic LLM workflows for imaging tasks—shipping reliable systems with measurable outcomes and strong observability.

I’m a Research Associate at IIT Jodhpur, working on automated detection, localisation, and anatomical characterisation of rib fractures in CT scans. My focus is on clinically interpretable subregion analysis and treatment-relevant outcome prediction.

I also investigate LLMs as reasoning modules within multi-agent clinical workflows—supporting automated radiology report generation and mistake-guided hierarchy learning for diagnostic classification. I’m especially interested in systems that improve both accuracy and diagnostic reliability.

Previously, I was a Data Science Intern at MIQ Digital, where I re-engineered a high-volume predictive pipeline in PySpark (migrated from R), reducing runtime by 67%. I also built GenAI-driven automation workflows using FastAPI and LLM orchestration for client-facing applications.

During my research internships, I developed FAIR-compliant deep learning models for cancer diagnosis and brain tumour segmentation and integrated reproducible models into the BioModels repository. Through my projects, I build agentic and neuro-symbolic frameworks—like LangGraph-based autonomous rib fracture workflows and mistake-aware hierarchy learning with critic–editor LLMs.

Experience

Work history, roles, and key accomplishments

Current

Research Associate

Current

IIT Jodhpur

Jul 2025 - Present (11 months)

Developing deep learning pipelines for automated detection, localisation, and anatomical characterisation of rib fractures in CT scans, enabling clinically interpretable subregion analysis and treatment-relevant outcome prediction. Investigating LLMs as reasoning modules in multi-agent clinical workflows, including radiology report generation and mistake-guided hierarchy learning for diagnostic cl

Deep Learning Rib Fracture Detection Anatomical Subregion Analysis Outcome Prediction Clinical Report Generation Mistake Guided Learning

Data Science Intern

MIQ Digital

Jan 2025 - Jun 2025 (5 months)

Re-engineered a high-volume predictive pipeline in PySpark (migrated from R), reducing runtime by 67%. Built GenAI-driven automation workflows for client-facing applications using FastAPI and LLM orchestration.

Pyspark Performance Optimization Migration From R fastAPI Orchestration Data Engineering

Research Intern

IIT Madras

May 2024 - Jul 2024 (2 months)

Developed and standardised 4 FAIR-compliant deep learning models for cancer diagnosis, brain tumour segmentation, and clinical risk prediction to support AI-assisted oncology treatment planning. Integrated the reproducible models into the BioModels repository while maintaining adherence to FAIR principles.

Deep Learning Medical Imaging Cancer Diagnosis Brain Tumor Segmentation Clinical Risk Prediction BioModels Integration Model Standardisation

Education

Degrees, certifications, and relevant coursework

Indian Institute of Science Education and Research, Bhopal

Bachelor of Science, Data Science and Engineering

2021 - 2025

Grade: 3.06/4.00 (CGPA)

Activities and societies: Submitted under-review research to NeurIPS 2026 (Fluid Hierarchies) and MICCAI 2026 (CLARIF) during the program.

Pursuing a B.S. in Data Science and Engineering at IISER Bhopal, completing coursework in data science, machine learning, and deep learning. Reported CGPA is 3.06/4.00.