# Arnav Deshpande

Arnav Deshpande is a remote worker on Himalayas, Machine Learning Intern at DRDO Research Centre Imarat, based in India.

Source: https://himalayas.app/@arnavdeshpande

```text
Machine Learning Engineer at DRDO, Research Centre Imarat, who improved evaluation quality and built tested LLM and agent pipelines.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | India |
| Authorized to work in | India |
| Salary expectations | 30k+ USD |
| Job categories | Machine Learning Engineer, Machine Learning Researcher |
| Skills | Machine Learning, LLM Systems, Model Training, Model Evaluation, Inference Benchmarking, RAG, Embeddings, Reranking, Agent Tools, Evaluation Suites, Fine Tuning, PEFT, DPO, MCP, GuardRails |
| LinkedIn | https://linkedin.com/in/arnav-deshpande-26a792290 |
| GitHub | https://github.com/Arnavdsp |

## About

```text
At DRDO, Research Centre Imarat (RCI) / DRDL, I improved an evaluation pipeline by replacing a scoring method that under-reported model quality with Normalized Wasserstein Distance. I also benchmarked three detectors across GPUs and traced contradictory dashboard metrics back to their data and evaluation sources.

As an Undergraduate Research Intern at IIT Indore, I exposed a false result in a synthetic-noise control that appeared to show embeddings tracking noise, then shipped the pipeline as a CLI package with 120 automated tests.

In my Evaluation Assurance project, I ranked evaluator verdicts using execution checks to catch incorrect PASS verdicts while sending only a portion of cases to human review. I also built an evaluation harness for EDGAR MCP agent tools, where ambiguous queries were refused correctly in graded runs and accuracy was reported as measured.

My projects include multi-agent decision support with human approval gates, deterministic verification, and an audit log, as well as hybrid RAG that abstains on unsupported answers and attributes responses to sources. I back my training, inference, RAG, and agent pipeline work with tests and written failure analysis.
```

## Experience

### Machine Learning Intern

DRDO Research Centre Imarat. May 2026 - Jun 2026

```text
Stopped an evaluation pipeline from under-reporting model quality 5x by proving IoU scoring failed on 5-15 px targets and replacing it with Normalized Wasserstein Distance. Benchmarked 3 detectors on accuracy and latency across A10G, L4 and T4 GPUs, and kept wrong conclusions out of review reports for 5 sub-projects.
```

Skills: Machine-Learning, Computer-Vision, Evaluation-Pipelines, IOU, Normalized-Wasserstein-Distance, Benchmarking, GPU, A10G, Cloud-Pipeline

### Undergraduate Research Intern

IIT Indore. May 2025 - Aug 2025

```text
Exposed a false result by running a synthetic-noise control that showed embeddings with a perfect ARI of 1.000 were tracking noise. Shipped the pipeline as a CLI package with 120 automated tests.
```

Skills: Unsupervised-Learning, Embeddings, ARI, Synthetic-Noise-Control, CLI, Testing-Automation

## Education

### Indian Institute of Technology Indore

B.Tech, Space Sciences and Engineering. 2023 - Present

```text
Pursuing a B.Tech in Space Sciences and Engineering (Engineering Physics) at IIT Indore, with coursework in machine learning, probability and statistics, optimization methods, and scientific computing.
```

### Anthropic

AI Fluency: Framework & Foundations, Artificial Intelligence

```text
Completed the AI Fluency: Framework & Foundations certification from Anthropic.
```

### Coursera

Deep Learning with PyTorch: Image Segmentation, Deep Learning

```text
Completed the Deep Learning with PyTorch: Image Segmentation course on Coursera.
```

### Coursera

Deep Learning Specialization, Deep Learning

```text
Completed the Deep Learning Specialization and Deep Learning with PyTorch: Image Segmentation courses on Coursera.
```

## Tech stack

- Google Analytics
- Python
- PyTorch
- scikit-learn
- FastAPI
- Linux
- pytest
- Git
- Docker
- GuardRails
- CUDA
- Google Cloud Run
- Hugging Face
- Qdrant
- LangChain
- LlamaIndex
- LangGraph
- PEFT
- Sentence Transformers

To contact Arnav, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Bapuray Patil](https://himalayas.app/@bapuraypatil): AI/ML Research Intern, DRDO (Cemilac)
- [Vaidehi User](https://himalayas.app/@vaidehi_amrutkar): Generative AI Intern, ARG Supply Tech Pvt. Ltd.
- [Aaditi Kumari](https://himalayas.app/@aaditikumari): LLM Post Training Intern, Ethara.AI
- [Prateek Singh](https://himalayas.app/@prateeksingh8): Research Intern, Indian Institute of Technology Ropar
- [Parth Jain](https://himalayas.app/@parthjain): Research Intern, Connectomics and Computational Biology, PRAN Foundation
- [Soumyajit Chakraborty](https://himalayas.app/@soumyajitchakraborty): Summer Research Intern, IIT Indore
- [Lochan Didla](https://himalayas.app/@lochandidla): AI Intern, ADAS Team, Daimler Truck Innovation Center India
- [Yash Chaudhary](https://himalayas.app/@yashchaudhary5): Data Science with GenAI Intern, Innomatics Research Labs
- [MANIK DABAS](https://himalayas.app/@manikdabas): Summer Research Intern, Defence Research and Development Organisation (DRDO)
- [CHIRAG DUA](https://himalayas.app/@chiragdua): Research Intern, DRDO – The Institute for Systems Studies & Analyses (ISSA)
