> Markdown version of [/jobs/ext/3804722-ai-ml-engineer](https://www.wearedevelopers.com/jobs/ext/3804722-ai-ml-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# AI/ML Engineer

- **Company:** SRC Inc.
- **Location:** Beavercreek Township, OH, United States
- **Salary:** $70,000.0 - $96,000.0
- **Contract:** Internship / Graduate position
- **Skills:** Artificial Intelligence, Cloud Computing, Computer Clusters, Python (Programming Language), Machine Learning, Language Modeling, NumPy, Open Source Technology, Reinforcement Learning, QLoRA, Pytorch, Evaluation Pipelines, Retrieval-Augmented Generation, Transfer Learning, Large Language Models, Deep Learning, Model Validation, Git, Pandas, Matplotlib, Containerization, Scikit Learn, Information Technology, HuggingFace, Machine Learning Operations, Evaluation of Large Language Models, Model Context Protocol, Docker
- **Published:** October 17, 2026
- **Apply:** https://www.indeed.com/viewjob?jk=b110591744e30bb4

## About the Role

* Bachelor's in computer science, Engineering, Mathematics, Physics, Statistics, or a related technical discipline and 0+ years of experience (internships, co-ops, research assistantships, and substantial academic project work count)
 * Working proficiency in Python and the core scientific stack (NumPy, pandas, scikit-learn, matplotlib)
 * Familiarity with at least one deep learning framework, preferably PyTorch
 * Exposure to the Hugging Face ecosystem (Transformers, Datasets, Tokenizers) or equivalent open-source model tooling
 * Understanding of ML fundamentals: train/validation/test discipline, overfitting and regularization, cross-validation, and the metrics appropriate to classification, regression, detection, and generation tasks
 * Comfort with Git-based collaborative development
 * Clear written and verbal communication, and a willingness to say when a result doesn't hold up

Ways to Stand Out

 * Experience fine-tuning open-weight language models using PEFT methods (LoRA, QLoRA) or the TRL/Axolotl/Unsloth toolchain
 * Hands-on work with RAG systems: embedding models, vector stores, chunking strategies, and retrieval evaluation
 * Familiarity with LLM evaluation practice - LLM-as-judge, rubric-based grading, eval harness design, and the failure modes of each
 * Experience with experiment tracking and data versioning tools (Weights & Biases, MLflow, DVC)
 * Exposure to reinforcement learning libraries (Gymnasium, Stable-Baselines3, CleanRL) or to preference-based post-training methods (DPO, GRPO)
 * Experience with agentic workflows, tool-calling, or the Model Context Protocol (MCP)
 * Work with non-natural-image modalities: RF, radar, SAR, audio, time series, or multimodal data
 * Containerization (Docker) and experience running jobs on GPU clusters or cloud compute
 * Master's degree in a related technical discipline

## Description

* Build, clean, label, and version datasets from real and simulated sources, including handling class imbalance, distribution shift, and sparse or noisy labels
 * Train, fine-tune, and benchmark models using standard open-source frameworks, running experiments as assigned and documenting results clearly
 * Develop and run evaluation pipelines - task-specific metrics, held-out and adversarial test sets, regression suites - and report findings honestly, including negative results
 * Support generative AI efforts through prompt development and iteration, structured output design, retrieval-augmented generation pipelines, and comparative evaluation of open-weight and commercial language models
 * Assist with parameter-efficient fine-tuning (LoRA/QLoRA) and supervised fine-tuning of open-weight models for domain-specific tasks
 * Implement reinforcement learning experiments in standard environments and instrument them for reproducible measurement
 * Maintain reproducible experiment tracking, environment configuration, and code hygiene so that results can be regenerated by others
 * Contribute data, figures, and written findings to technical reports, program reviews, and oral presentations
 * Collaborate with algorithm engineers and domain experts to understand the sensor data and physical phenomenology behind the datasets you work with