Senior AI/ML Engineering Architect — TS/SCI w/ Polygraph (Hybrid)
Posted
Summary
Senior AI/ML architect at Finch AI who leads the evaluation, fine-tuning, and production deployment of LLMs/SLMs, building agentic AI, RAG, and NLP text-analytics systems shipped into classified federal customer environments. Hybrid role in Herndon, VA requiring an active TS/SCI with polygraph; core stack is Python plus LangGraph, PyTorch, Hugging Face, spaCy, and scikit-learn.
Senior AI/ML Engineering Architect — TS/SCI w/ Polygraph
Location: Hybrid, Herndon, VA (Required)
Clearance: Active TS/SCI w/ Polygraph
Compensation: $200,000 – $240,000 base salary
About Finch AI
Finch AI exists because you can't miss. We build AI-powered solutions that transform raw information into decisive action. We deliver accurate, trusted, real-time insights from data so that analysts, operators, and executives can understand risk and act with confidence, even when the data is massive, complex, and time-sensitive.
Our platform serves organizations across national security, research security, defense, finance, law enforcement, and legal sectors. We don't start with our technology — we start with your challenge. Our integrated approach combines a powerful platform, deep domain expertise, and the flexibility to meet customers where they are.
Finch AI holds 30+ patents and has had products deployed to government agencies for over 15 years. Our team spans former analysts, PhD data scientists & AI engineers, and commercial data specialists who have developed products and capabilities at some of the largest tech companies and agencies in the world.
Our culture:
We operate as a mission-flexible organization where integrity is at the center of everything we do. We value sharp questions, challenged assumptions, and cross-disciplinary collaboration. No bureaucracy stands between great ideas and real impact — ownership is trusted, expertise is shared freely, and the work genuinely matters.
We're looking for people who are energized by complex, high-stakes problems; who care about accuracy, transparency, and building technology people can trust; and who communicate with clarity and curiosity in a distributed environment. If that sounds like you, we'd love to talk.
The Role
The systems you'll build here are the engine behind that mission and in this role, they'll run in some of the most demanding environments in the federal government. You'll join the team building Finch AI's next generation of products, helping design and build systems behind them end to end — from design through production deployment in cleared customer environments and across the broader business. As a technical leader in generative AI and large language models, you'll help set direction on how LLMs and SLMs are applied across the product line.
Responsibilities
- Lead the evaluation, fine-tuning, and deployment of large language models (LLMs) and small language models (SLMs) in production applications.
- Build AI-enabled applications using agentic AI, retrieval-augmented generation (RAG), and prompt engineering.
- Design, develop, and evaluate machine learning models for text analytics — extraction, disambiguation, classification — including custom algorithm development where off-the-shelf approaches fall short.
- Carry solution designs from proof of concept and prototype through to shipped features in existing products.
- Collaborate with customer teams to deliver mission-aligned AI solutions within classified environments.
- Partner with product teams and stakeholders in an Agile environment, communicating clearly with technical and non-technical audiences.
Requirements
- Hands-on experience building and deploying LLM and SLM applications: agentic AI, RAG, prompt engineering, and fine-tuning.
- Production experience with NLP systems for data classification, entity extraction, and information extraction, covering the full lifecycle — collecting and preparing training data, training supervised and unsupervised models, tuning hyperparameters, and analyzing model errors.
- Applied deep learning for NLP, plus strong working knowledge of Bayesian statistics, statistical analysis, and predictive modeling.
- Strong Python skills; a second language such as Java, Go, or C++ is a plus.
- Working familiarity with LangGraph, PyTorch, Hugging Face Transformers, spaCy, scikit-learn, NumPy, Pandas, and Jupyter.
- Experience with SQL, NoSQL, and graph databases.
- 5+ years of experience building commercial software products.
- MS or PhD in Computer Science or a related field, or equivalent practical experience (Bachelor's degree minimum).
- Active TS/SCI w/ Polygraph
Finch AI is an equal opportunity employer. Reasonable accommodations are available during the hiring process by contacting hr@finchai.com.