AI Software Engineer

**About the Role** BSL is seeking a technically proficient AI Software Engineer to join our AI team. The role sits at the intersection of machine learning, agentic systems, and architectural technology (AEC). You will develop ML models that interpret complex architectural documents and agentic systems that enable click\-to\-takeoff, inventory inquiry, order placement and tracking. **Key Responsibilities** * Design and iterate on prompts, context, and model selection for production tasks. * Build the tools and integrations agents call, and the retrieval layer that grounds them. * Orchestrate multi\-step agent workflows, including error handling, retries, and human escalation. * Develop evaluation sets and measure accuracy, cost, and latency to support evidence\-based iteration. * Ship production application and API code. * Build responsive interfaces in React and Tailwind CSS and manage deployments on Vercel and other cloud platforms. **Required Qualifications** * Enrolled in or recently graduated from a BS/MS program in Computer Science, Electrical Engineering, Applied Math, or a related field. * Solid grounding in ML fundamentals: model training, evaluation, generalization, and error analysis. * Practical experience with PyTorch or TensorFlow. * Ability to design experiments, quantify results with appropriate metrics, and communicate findings to a non\-ML audience. * Production\-level use of LLM APIs (Anthropic, OpenAI, or equivalent) via SDK or REST. * Proficiency in prompt engineering, tool/function calling, and structured output. * Experience implementing multi\-step LLM workflows with failure handling and fallback logic. * Strong Python, including async, type hints, and testing. Proficient with Git. * Deployment experience on Vercel, AWS, GCP, or Azure. * Demonstrated record of shipping and maintaining software used in production. **Preferred Qualifications** * Agent frameworks and protocols: Claude Agent SDK, OpenAI Agents SDK, LangGraph, LlamaIndex, MCP. * Retrieval and vector search: pgvector, Qdrant, Pinecone; chunking and reranking strategies. * Evaluation and observability: LangSmith, Braintrust, Langfuse, ragas, or a custom harness. * Data validation: Pydantic, JSON Schema. * Multimodal and document AI: vision language models, OCR, PDF parsing, layout analysis, object detection. * Infrastructure: Docker, CI/CD, PostgreSQL. * Domain: background or demonstrated interest in Architecture, Engineering, and Construction (AEC), including familiarity with floor plans and technical naming conventions. **What You'll Gain** * Hands\-on experience in AI\-powered automation and ERP integration. * Potential for full\-time employment based on performance. **Application Requirements** * Resume/CV * GitHub profile * Academic transcript BSL is an equal opportunity employer. We consider all qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other protected characteristic. Pay: $20\.00 \- $35\.00 per hour Expected hours: 40\.0 per week Benefits: * Flexible schedule Work Location: In person