AI/ML Engineer

About Saltmine

• Saltmine is building an AI-powered workplace strategy and capital projects platform for
• enterprise real estate teams. We help global organizations move from business goals →
• spatial programs → stack plans → test fits → execution in a seamless, data-driven
• workflow.

Our platform combines

• ● Workplace strategy
• ● Occupancy planning
• ● Capital project planning
• ● AI-driven automation
• to enable faster, smarter, and more cost-efficient real estate decisions.
• We are now building a next-generation automated test-fitting engine that can generate
• high-quality, architecturally valid layouts in minutes and we’re looking for a Spatial Algorithms
• Engineer to help power this core capability.

Role Overview

• As a Spatial Algorithms Engineer, you will design and build the core layout engine that
• generates optimized floor plans based on spatial constraints, architectural rules, and business
• requirements.

You will work at the intersection of

• ● computational geometry
• ● optimization algorithms
• ● real-world workplace design
• Your work will directly power Saltmine’s ability to automatically generate and evaluate
• thousands of layout options for enterprise customers.

Key Responsibilities

1. Build Automated Test-Fit Engine

● Develop algorithms to generate 2D layouts within complex floor plates

● Solve for

• ○ space allocation
• ○ corridor generation
• ○ adjacency constraints
• ○ geometric validity

1. Design Constraint-Based Systems

● Encode real-world rules into the system

• ○ door placement
• ○ circulation access
• ○ daylight constraints
• ○ minimum room geometry
• ● Implement hard and soft constraint frameworks

1. Optimization & Layout Generation Systems

● Build multi-objective optimization systems to balance

• ○ adjacency satisfaction
• ○ space efficiency
• ○ daylight utilization
• ○ constraint compliance

● Apply techniques like

• ○ greedy algorithms
• ○ simulated annealing
• ○ genetic algorithms
• ○ constraint solvers

1. Geometry & Spatial Modeling Systems

● Work with

• ○ polygons and floor plate boundaries
• ○ spatial partitioning
• ○ graph-based circulation systems

● Implement

• ○ space subdivision
• ○ packing algorithms
• ○ layout refinement logic

1. Scoring & Evaluation Systems

● Design scoring engines to evaluate layout quality

• ○ adjacency score
• ○ circulation efficiency
• ○ residual space
• ○ program compliance

1. Performance Optimization

● Ensure system can

• ○ generate multiple layouts efficiently
• ○ scale to large enterprise floor plates
• ○ support near real-time feedback

1. Cross-Functional Collaboration

● Work closely with

• ○ product managers
• ○ architects / workplace strategists
• ○ frontend / visualization teams

Required Skills

Core Computer Science & Algorithms

● Strong foundation in

• ○ data structures
• ○ algorithms
• ○ graph theory

Computational Geometry

● Hands-on experience with

• ○ polygon operations (intersection, clipping, partitioning)
• ○ spatial reasoning and layout systems

● Familiarity with libraries like

• ○ Shapely / GEOS
• ○ JTS / Turf.js
• ○ CGAL

Optimization Techniques

● Experience with

• ○ constraint solving
• ○ heuristic optimization

● Exposure to

• ○ simulated annealing
• ○ genetic algorithms
• ○ linear/integer programming
• Note: We are open for any other technique, the idea is to work towards a solution which is best
• in industry.

Programming

● Strong in

• ○ Python (preferred)
• ○ or C++ / Rust for performance-critical systems

Data Structures for Spatial Systems

• ● Grids, graphs, spatial indexing
• ● Quadtrees / BSP trees (good to have)

Good to Have

● Exposure to AEC / architecture / workplace planning

● Experience with

• ○ CAD / BIM tools (AutoCAD, Revit)
• ○ Rhino / Grasshopper

● Background in

• ○ game development (procedural generation)
• ○ robotics (path planning)
• ○ simulation systems

What We’re Looking For

• ● Strong problem-solving mindset with ability to translate real-world constraints into
• algorithms
• ● Comfort working with geometry, layouts, and visual systems
• ● Ability to balance multiple competing constraints in optimization problems
• ● Ownership mindset and ability to build 0 → 1 systems

Impact

• ● Build the core engine powering automated workplace planning
• ● Directly influence how global enterprises design offices
• ● Work on a category-defining product in PropTech + AI

Interview Process

• ● Algorithm & case study round - A real problem statement
• ● Discussion around case study
• ● Technical discussion
• ● Final discussion with leadership