Hindustan Pressings Private Limited

Hiring: Founding AI Engineer (Manufacturing Intelligence) — Stealth Build If you can read a STEP file and tell a factory what to do—with code—you’re who I’m looking for. What you need to know Build CAD ingestion + feature recognition for sheet metal parts (STEP/Parasolid/JT), including thickness, bends, cut paths, hole patterns, and weld seams. Design a rules/ML engine that converts geometry into manufacturable process plans (laser/press-brake/weld/stamping) with cycle-time and throughput estimates. Ingest and model time-series data on materials, labor, energy, logistics, and shop productivity to drive region-aware recommendations. Ship a production API and a lightweight web UI (upload → instant plan → clear explainability). Establish data quality, auditability, and versioning from day one. You should have 4–8 years building production ML/AI systems (Python preferred) AND strong software engineering fundamentals (testing, CI/CD, profiling). Hands-on with geometric kernels/tooling (Open Cascade, CGAL, trimesh, mesh/solid ops) or equivalent real experience. Manufacturing/IE intuition: cycle-time modeling for laser, bending, welding, stamping; fixture/setup thinking; OEE basics. Data plumbing: APIs/ETL, Postgres/Timeseries, caching, and basic MLOps. Ability to turn ambiguity into shipped code without hand-holding. Nice to have Experience inside a fab shop or on a production floor. Knowledge of PLM/ERP integrations, BOM/route structures. Go/TypeScript/Next.js for services and UI. Stacks Python (FastAPI), Postgres/Timescale, S3, Open Cascade/trimesh, Next.js. Cloud-agnostic. Location Pune preferred; remote (India) for the right person. Full-time. How to apply DM me on LinkedIn with: Links or brief write-ups for two projects: Parsing/working with CAD/geometry, or Turning engineering rules into a planner/scheduler/estimator, or Building a production ML service with clear SLAs. Details about CTC will be discussed on call or in person.