AI/ML Engineer AI/ML · Boston (Preferred) or Remote · Full Time

xFigura; Boston (MA); Hybrid; USD 180,000 - 240,000; Full time

Benefits offered by this job

Equity

Health insurance

Dental

Vision

Flexible PTO

High ownership

Job summary

xFigura in Boston is hiring a platform engineer to own end-to-end AI orchestration, retrieval and knowledge systems, data pipelines, and collaboration features for an AI-native design platform used by leading architects and universities.

You will ship core systems from concept to deployment, scale performance and infrastructure, and partner with leaders to shape the architecture and product roadmap while maintaining high ownership and autonomy.

Qualifications

• Track record shipping full production systems end-to-end.
• Production experience with AI/ML systems — LLM/generative pipelines, retrieval, or model serving.
• Working knowledge of data engineering fundamentals: pipelines, storage, transformation, and data quality.
• Highly effective use of AI development tools.
• Comfort operating with autonomy and ambiguity.
• Strong problem-solving and communication skills.

Responsibilities

• Design, build, and ship features and systems end-to-end
• Architect scalable frontend and backend infrastructure
• Build and own AI orchestration, RAG, and agent pipelines in production
• Build the data pipelines and ingestion infrastructure those systems depend on
• Develop canvas tooling and integrations with professional design software
• Work directly with users to translate architectural workflows into systems
• Drive technical decisions that shape how the platform evolves
• Raise engineering standards and help interview and level new engineers
• Ship fast, measure, and iterate

Job description

You'll work closely with the leadership team, including the Head of R&D, to build the core systems behind an AI-native design platform used by leading architecture firms, designers, and universities.

This is foundational work: orchestrating generative models, building the retrieval and data infrastructure that grounds them in real project knowledge, powering collaborative canvas tooling, and integrating with the CAD ecosystem architects already live in. We have production usage and enterprise adoption today. The next phase is scaling capability, performance, and infrastructure — and you'll own major systems that the rest of the platform depends on.

You'll ship systems end-to-end, from concept through deployment, translating real architectural workflows into software while raising the technical bar across the team.

What You'll Work On

AI orchestration — Routing, chaining, and evaluating 50+ generative models (image, geometry, text) behind a single coherent product surface. Model selection, fallback strategy, cost/latency tradeoffs, streaming inference, and GPU queue management.

Retrieval & knowledge systems — RAG over enterprise design knowledge: document and drawing ingestion, chunking and embedding strategy, vector and hybrid (semantic + lexical) search, re-ranking, and agentic retrieval where the model decides what to look up and when. Making retrieval accurate enough that architects trust the output.

Data engineering & pipelines — Batch and streaming pipelines that turn messy CAD files, PDFs, specs, and project archives into structured, queryable, embeddable data. Schema design, incremental processing, lineage, and the observability to know when a pipeline silently degrades.

Evaluation & reliability — Eval harnesses, regression suites, and tracing for non-deterministic systems. Knowing whether a prompt, model, or retrieval change actually made the product better.

Infinite canvas collaboration — Real-time multiplayer editing, conflict resolution, and creative tooling that stays responsive at scale.

CAD & BIM integrations — Rhino, Speckle, and Revit. Geometry interchange, plugin surfaces, and round-tripping data without losing fidelity.

Responsibilities

Design, build, and ship features and systems end-to-end

Architect scalable frontend and backend infrastructure

Build and own AI orchestration, RAG, and agent pipelines in production

Build the data pipelines and ingestion infrastructure those systems depend on

Develop canvas tooling and integrations with professional design software

Work directly with users to translate architectural workflows into systems

Drive technical decisions that shape how the platform evolves

Raise engineering standards and help interview and level new engineers

Ship fast, measure, and iterate

Required Qualifications

Track record shipping full production systems end-to-end, not just features

Production experience with AI/ML systems — LLM/generative pipelines, retrieval, or model serving

Working knowledge of data engineering fundamentals: pipelines, storage, transformation, and data quality

Highly effective use of AI development tools

Comfort operating with autonomy and ambiguity

Strong problem-solving and communication skills

Preferred Qualifications

Real-time or collaborative systems experience (CRDTs, OT, WebSockets)

Canvas, node-based, or graph UI familiarity

Vector databases, embedding models, or search relevance tuning

LLM evaluation, tracing, or observability tooling

Orchestration frameworks and workflow engines (Airflow, Dagster, Temporal, or similar)

Compute pipeline and API integration experience

Exposure to CAD/BIM or AEC workflows

Cloud and GPU infrastructure experience

Early-stage startup background

Competitive salary plus meaningful equity

Health, dental, and vision

Flexible PTO

High ownership and autonomy

What You'll Gain

Ownership of foundational platform systems

Close collaboration with engineering, product, and R&D leadership

Real influence over architecture and roadmap