Phone Number: 732 966 5584 | Location: Toms River, NJ | Email: wbaltus21@gmail.com | Linkedin: www.linkedin.com/in/william-baltus
Designed and implemented advanced AI/ML, full-stack, and automation systems to modernize enterprise operations and written space-hardened aerospace software. Developed a production-scale Retrieval-Augmented Generation (RAG) platform leveraging Elasticsearch (BM25, dense/sparse vectorization) and Large Language Models (LLMs) to enhance retrieval accuracy across 1M+ classified records. Wrote Class B flight software for the LM400 satellite line using Python and C++, conducting rigorous unit, integration, and simulation-based testing with hardware replicas to meet aerospace compliance standards. Advanced enterprise payroll automation and compliance efforts by modernizing legacy HCM validation workflows into AI-enabled, data-driven systems —supporting migration from legacy platforms to scalable, intelligent architectures across Lockheed’s enterprise ecosystem.
Stevens Institute of Technology - Hoboken, NJ
Bachelor of Engineering in Software Engineering - 2022
Graduated with 3.95 GPA
Duke University - Durham, NC
Master of Science in Electrical and Computer Engineering - 2025
Software Engineer Sr. | HCM Lockheed Martin, Space - Jan 2024 to present
- Modernized Lockheed Martin’s legacy Payroll Validation System into a scalable, automated data validation platform, significantly reducing manual efforts and contributing to an estimated $8M in annual savings.
- Engineered core framework components across the UI, API, ingestion pipeline, and PostgreSQL data layer within a Python-based Fast API MVC architecture, automating ingestion, validation, and archival for SAP-integrated workflows pertaining to enterprise payroll.
- Implemented observability and performance tracking tools to surface validation metrics, ingestion status, and audit readiness for engineering and business teams.
- Contributed to an AI + MCP (Multi-Component Protocol) proof-of-concept, demonstrating the potential for AI-assisted rule generation and next-phase validation automation.
- Partnered closely with senior engineers, analysts, and HCM stakeholders to transition manual validation workflows into a reusable automation framework, improving reliability, scalability, and governance across HR systems.
AI/ML Engineer | Savant Lockheed Martin, Space - Jan 2024 to Jan 2025
- Engineered the backend and UI between IBM Watson AI–powered document retrieval and Savant using C# and .NET, integrating Blue Prism RPA workflows and Python automation scripts to streamline data synchronization and scheduled task execution.
- Contributed to an enterprise-level Retrieval-Augmented Generation (RAG) system that enhanced search across 1+ million records.
- Optimized an Elasticsearch cluster utilizing BM25, Dense, and Sparse Vectorization techniques.
- Leveraged pretrained Large Language Models (LLMs) and Transformer models from Hugging Face to create a private RAG-as-a-Service (RAGaaS).
- Enhanced hybrid search functionality with Reciprocal Rank Fusion (RRF) algorithm in Elasticsearch, ensuring semantic relevancy of retrieved documents for LLM processing.
- Conducted scoring evaluation and reranking to fine-tune the RAG system, achieving highly accurate responses.
Software Engineer | SSPL/Mustang Lockheed Martin, Space - Sept 2022 to Jan 2024
- Employed technical expertise in Python and C++ development to implement object-oriented principles in a collaborative and fast-paced environment.
- Successfully increased team velocity from .75 to 2.90 as scrum master over the course of 6 months to improve program efficiency and produce on time delivery of code.
- Developed, tested, and verified Space Software Product Line software systems built on top of NASA cFS with a software factory mentality to deliver LM400 satellites.
- Managed a backlog filled with Cameo work, assigned tickets to teammates, and provided assistance to associate engineers to translate FSW to diagrams.
Cross System Software IPT Lead Lockheed Martin, Space- ELDP - Jun 2023 to July 2024
- Acted as Software IPT Lead across 3 subsystems to deliver a demo lunar tube exploration vehicle and satellite tracker. POC for SW deliverables for SDR, PDR, CDR, and TRR.
- Led software and system architecture discussions to produce INAR and Proposal documents based on SOW and RFP to win contract award.
- Decomposed SOW to actionable features and created subsystem architectures. Distributed work amongst software engineers. Produced initial implementations to kickstart each team.
- Developed and integrated simultaneous localization and mapping (SLAM), autonomous navigation algorithms (Dijkstra), and hardware interface for vehicle. Using ROS2, C++, Python, Ubuntu MATE.
Programming/Scripting Languages:
Python | JS/TS | C/C++ | Java| C#| LABVIEW| Bash | PowerShell | Luau
AI Frameworks:
Langchain/Langgraph| LlamaIndex | Haystack| MCP| A2A|
Frontend Frameworks:
React | Angular | Flask | Django| Streamlit| Vue.js| Remix.js
Backend Frameworks:
.NET Core | Node.js| FastAPI
Automation Frameworks:
Blue Prism | Power Automate | Selenium | Robocorp | Beautiful Soup
Data & Vector Databases:
ElasticSearch | PostgreSQL | MongoDB | MySQL | SQL Server| RabbitMQ | S3
Deployment/Hosting:
Docker | OpenShift | Firebase| AWS| GCP| Heroku
Robot Frameworks:
ROS2