AI Engineer - Direct Hire Only, no Corp-2-Corp

US citizenship is required for this position. This requirement is imposed by the terms of the federal contract under which this work is performed and by federal agency access requirements, and it is not a Pistevo preference. The selected candidate must pass a federal background investigation and receive an agency suitability determination prior to starting work on the program.

This is a direct-hire W-2 position. We do not accept corp-to-corp arrangements, 1099 contractors, or submissions from third-party recruiting agencies.

About Pistevo Decision

Pistevo Decision is a technology firm building AI and data platforms for federal agencies and commercial clients. We turn complex data into decisions people can act on, across defense, health, civilian missions, and other industry sectors.

We are small enough that your work ships and large enough that it matters. There is no layer of process between what you build and what a customer uses. We hire people who want to learn quickly, ask direct questions, and take ownership of something real.

We back that with flexible hours, remote and hybrid work where the contract allows it, comprehensive health insurance, generous PTO, and a real budget for training, certifications, and conferences. You will be mentored, not thrown in. Every junior engineer here works alongside a senior who reviews their code and answers their questions.

About the role

We built a working AI prototype for an IRS program. Your job is to help turn it into a maintained production platform.

This is a build-and-operate role, not a research role. You will write application code, wire up cloud services, fix things that break, and write the documentation that keeps a system auditable. You will be working directly with a team of engineers.

We are deliberately hiring at the junior level. We expect you to be early in your career, and we have structured the role so you are never the only person responsible for a production system.

What you will do

• Build and extend features in the platform's Python backend, from API endpoints through to data models

• Integrate large language models into application workflows: prompt design, retrieval, evaluation of output quality, and handling of failure cases

• Build and maintain retrieval pipelines: document ingestion, chunking, embedding, and indexing

• Write tests. We hold a high automated coverage bar, and it is not negotiable

• Work with our infrastructure-as-code and CI/CD pipelines to deploy changes safely to dev and production environments

• Add structured logging, monitoring, and alerting so we know when something is wrong before the customer does

• Write clear technical documentation, including the artifacts federal programs require for security and compliance review

• Participate in code review, both giving and receiving it

Technology you will work with

• Language and framework: Python/Django for backend, TypeScript/Angular for frontend

• AI/ML: Amazon Bedrock, retrieval-augmented generation, vector and graph-based retrieval, prompt engineering and evaluation

• Cloud: AWS, including Lambda, Step Functions, ECS Fargate, S3, Aurora PostgreSQL, and CloudWatch. Federal workloads run in AWS GovCloud

• Data: PostgreSQL, plus graph data stores for knowledge retrieval

• Infrastructure and delivery: Docker, Terraform, GitHub Actions

• Quality tooling: Ruff, pytest, coverage.py, pre-commit, type checking with mypy

Minimum qualifications

• Bachelor's degree in computer science, engineering, data science, or a related field, or equivalent practical experience

• 0–2 years of professional software engineering experience. Internships, co-ops, and substantial personal or academic projects count

• Working proficiency in Python and TypeScript

• Familiarity with web frameworks and with relational databases and SQL

• Hands-on experience calling LLM APIs and building something with them, whether at work, at school, or on your own

• Comfort with Git and collaborative development

• Ability to write clearly

• Willingness to work under a formal change control and documentation process

Preferred qualifications

• Exposure to AWS or another major cloud provider

• Experience with retrieval-augmented generation, vector databases, or embedding models

• Experience with Docker, Terraform, or CI/CD pipelines

• Coursework or hands-on work in NLP or machine learning

• Any prior exposure to federal contracting, FISMA, NIST 800-53, or IRS Publication 1075

• AWS certification, at any level

Pistevo Decision LLC is an equal opportunity employer. We consider all qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, protected veteran status, or any other characteristic protected by law.

Pay: $76,972.96 - $101,784.64 per year

Benefits:

• 401(k)

• Dental insurance

• Flexible schedule

• Flexible spending account

• Health insurance

• Health savings account

• Paid time off

• Professional development assistance

• Retirement plan

• Vision insurance

Education:

• Bachelor's (Required)

Experience:

• Software design: 1 year (Required)

Ability to Commute:

• Herndon, VA 20171 (Preferred)

Work Location: Hybrid remote in Herndon, VA 20171