Machine Learning & Data Engineer

Machine Learning & Data Engineer - Innovation Division

Dynamo Technologies is seeking a hybrid Machine Learning and Data Engineer to join our Innovation Division, a dynamic research and development team focused on rapidly building functional applications that deliver measurable efficiency gains for our clients and internal tools so our existing team can accomplish more. This role is for builders who thrive at the intersection of cutting‑edge AI technology and practical implementation, working in a fast‑paced environment where speed of delivery and thoughtful execution are paramount.

You’ll be responsible for developing intelligent systems that transform unstructured data into actionable insights, building pipelines that scale across cloud and self‑hosted environments, and helping clients understand how agentic AI systems can be deployed securely and cost‑effectively in federal environments.

In this role, you’ll work across teams to rapidly prototype and deploy AI‑powered solutions. Your day‑to‑day will involve extracting data from complex documents using multimodal AI models and OCR tools, building data pipelines that convert PDFs and unstructured documents into structured, normalized formats, and implementing PII and IP redaction systems for sensitive applications. You’ll design and deploy retrieval‑augmented generation systems for chatbots and knowledge‑management platforms, integrating graph databases and vector stores to enhance document retrieval and question‑answering capabilities.

You’ll also be presenting your work to both internal stakeholders and external clients, demonstrating software capabilities and helping shape solution architectures for federal contracting opportunities. This means understanding the RFP process, solution development, and how to communicate technical capabilities in ways that resonate with decision‑makers.

• Demonstrated experience with text extraction using multimodal AI models and OCR tools for processing multi‑page applications and forms
• Proven ability to build data pipelines on both AWS and self‑hosted Linux environments that transform unstructured data from PDFs and documents into structured, normalized formats
• Experience implementing PII redaction and name/IP redaction systems
• Strong presentation skills with a track record of presenting software and technical solutions to stakeholders and leaders for internal and external R&D projects
• Understanding of federal contracting environments, RFP processes, and solution architecture development

• Experience with hybrid database architectures, particularly Neo4j graph databases alongside vector stores like Qdrant and Pinecone
• Working knowledge of retrieval‑augmented generation (RAG) for chatbot and knowledge‑management system development
• Hands‑on experience deploying agentic systems including LLMs, image generators, image classifiers, and video generators, with understanding of deployment patterns, sizing and scaling considerations, and token architecture
• Ability to design intelligent systems that integrate seamlessly with each other across platforms
• Active security clearance for defense work
• Bachelor’s or Master’s degree in Computer Science, Data Science, Engineering, or related field

• Proficient in Python, SQL, Git, and Docker
• Cloud‑agnostic and platform‑agnostic mindset with ability to work across different environments
• Helpful:
  Experience with front‑end application development using frameworks such as Tailwind, JavaScript, Svelte/Svelte Kit, React, Angular, or similar JavaScript frameworks
• Comfortable building and configuring hardware and infrastructure when needed

Bachelor’s degree in Computer Science or 5–10 years of equivalent work experience

You’re a builder and maker at heart who doesn’t mind getting hands‑on with infrastructure and hardware. You’re someone who gets excited about self‑hosted agentic systems and can articulate why they’re more secure and cost‑effective for federal clients. You’re comfortable learning new platforms quickly and aren’t afraid to switch between tools to find the best solution for each…