Software Engineer – AI Applications (RAG, LangChain, Vector DBs)

Boast AI platform brings visibility into how R&D teams operate and allocate their R&D time and resources to prepare the largest R&D Tax Credit with the least distraction to the engineering teams. We connect engineering activity with financial data to surface insights into cost, productivity, and investment ROI. We use state-of-the-art AI techniques to summarize work, tag themes, map skills, forecast costs, and deliver insights via interactive dashboards and conversational interfaces.

About the Role

We’re looking for an engineer who’s not just familiar with building AI features — but can own the backend AI pipelines, leverage RAG architecture, and integrate LLMs into a production environment using tools like LangChain and pgvector.

What You'll Do

• Design and implement Retrieval-Augmented Generation (RAG) pipelines that combine structured and unstructured data from multiple sources.
• Build LLM-powered APIs and agents using LangChain, LlamaIndex, or similar frameworks.
• Work with vector databases like pgvector, Weaviate, or Pinecone to store and retrieve document embeddings.
• Ingest and process data from tools like Jira, GitHub, GitLab, Confluence, payroll systems, and more.
• Generate work summaries, auto-tag themes, and map skills using LLMs and embeddings.
• Train or fine-tune lightweight models for text classification, summarization, and embedding generation when needed.
• Develop fast, scalable backend services in Python (FastAPI preferred) to expose AI insights via REST or GraphQL APIs.
• Collaborate with product and engineering leadership to define and deliver high-impact AI-powered features.

Who You Are

• You’ve built or deployed RAG systems in production, combining LLMs with your own document/data stores.
• You’re comfortable building with LangChain, LlamaIndex, or similar orchestration frameworks.
• You have hands-on experience with vector databases like pgvector, FAISS, Weaviate, or Pinecone.
• You can write robust, maintainable Python code, ideally with FastAPI, Airflow, or similar frameworks.
• You understand prompt engineering and can work with OpenAI, Gemini, or open-source LLMs (Llama, Mistral, etc.).
• You’ve worked with embedding models and understand how to choose or fine-tune models for different NLP tasks.
• You’re pragmatic — balancing experimentation with getting working features into the hands of users.
• Bonus: Experience with GCP infrastructure, including Cloud Functions, and Vertex AI.

Tech Stack You'll Work With

• Python, Airflow, Node.js
• PostgreSQL, pgvector, MongoDB
• LangChain, OpenAI, Gemini,
• Fivetran for data ingestion
• Typescript, React

Why You'll Love Working at Boast

• Competitive base salary of $125K-$145K, plus a robust benefits package.
• Hybrid and flexible work culture.
• You’ll work directly with experienced product and technology leaders.
• You’ll own foundational AI systems powering a real product, not a sandbox experiment.
• You’ll be part of a fast-moving team focused on building useful, high-impact tools for modern R&D organizations.
• We embrace AI as part of how we work, not just what we build.