Duckdb Engineer Jobs in Boston, MA (NOW HIRING)

Your Impact at LILA

We're building an AI-driven drug discovery factory that closes the loop between computational design and wet-lab experiment in days instead of months. The bottleneck for that loop is data: every compound designed, synthesized, QC'd, and tested needs to be registered, tracked, and surfaced back to our AI Scientists with low latency and high fidelity.

As the Senior Engineer for the Drug Discovery Platform, you will own the data backbone that makes this closed loop possible: the molecule queue, the compound registration system, the synthesis constraints, and the experimental result capture pipelines that feed back into our predictive models. Your work will directly determine how fast our AI learns from each cycle, and therefore how fast we deliver new medicines.

What You'll Be Building

• Molecule Queue: Design and operate the shared queue that brokers compounds between AI Scientists and the Make-Test platform, including the status state machine, batch grouping, priority semantics, and the read/write contracts both sides depend on.
• Compound & Structure Registration: Build the canonical molecule registry, including SMILES/InChI normalization, stereochemistry handling, salt/parent resolution, duplicate detection, and stable internal IDs (e.g., LILA-NNN) that every downstream system can rely on.
• Lab Instrument Integration: Build pipelines that ingest data from ChemSpeed automated synthesis, QC instruments (LCMS, NMR), and bioassay readers, capturing identity, purity, dose-response curves, IC50/EC50, ADMET measurements, and selectivity panels into a queryable store.
• Synthesis Constraints & Inventory: Maintain the live state that the Batch Assembly AI reads each cycle, covering building-block inventory, advanced precursors, ChemSpeed capacity, stock alerts, and chemistry-specific constraints (reaction types, step budgets, solvent compatibility).
• Reliability & Observability: Own data quality, lineage, schema evolution, and SLAs for the closed-loop cycle time target of a few days from computational proposal to experimental truth.

What You'll Need to Succeed

• Education & Experience: Bachelor's or Master's in Computer Science, Chemistry, Computational Biology, or a related field, and 5+ years building data platforms in production, ideally including some exposure to scientific or lab-generated data.
• Platform Engineering: Designed and shipped data platform components from the ground up (ingestion frameworks, registries, storage abstractions, and orchestration); fluent in backend production APIs/services, Python and SQL and writes production-quality code.
• Database & Schema Design: Production experience with relational and/or NoSQL databases, schema design for evolving scientific data, and query optimization; comfortable across structured, semi-structured, and unstructured data.
• Lab/Scientific Data Fluency: Comfortable modeling chemical and biological data (structures, reactions, assays, dose-response, batches), and the messy reality of experimental measurements (replicates, censored values, failed runs).
• Cloud & Infrastructure: Experience with AWS and containerized deployment (Kubernetes).
• AI-Assisted Development: Proficient with AI-assisted development tools (Cursor, Claude Code, or similar) and incorporates them effectively into day-to-day engineering work.

Bonus Points For

• Cheminformatics: Hands-on with RDKit, OpenEye, or equivalent (structure standardization, registration, and search).
• Compound Registration Systems: Direct experience building or operating a corporate compound registry (CDD Vault, Dotmatics, Benchling Registry, or similar), or building one from scratch.
• ELN/LIMS Integration: Familiarity with electronic lab notebooks, LIMS, and instrument data formats (mzML, AnIML, vendor-specific).
• Workflow Orchestration: Experience with Flyte, Airflow, Dagster, or Temporal, especially for long-running scientific pipelines.
• Lakehouse Patterns: Modern table formats (Iceberg, Delta Lake, Hudi) and columnar processing (DuckDB, Polars, Spark).
• Automation/Robotics Data: Experience capturing data from automated synthesis platforms (ChemSpeed, Chemspeed SWING, Opentrons) or HTS readers.
• Domain Background: Prior work in pharma, biotech, or academic drug discovery; familiarity with SAR, ADMET, IC50/EC50, and DEL screening.
• Agentic/LLM Workflows: Experience building data infrastructure that serves agentic and LLM-driven workflows, with an understanding of the retrieval, latency, and grounding constraints automated systems impose.