Data Scientist (Dallas)

Codvo.ai•Dallas, TX

38d•Onsite

About The Position

At Codvo, we are committed to building scalable, future-ready data platforms that power business impact. We believe in a culture of innovation, collaboration, and growth, where engineers can experiment, learn, and thrive. Join us to be part of a team that solves complex data challenges with creativity and cutting-edge technology. The on-site technical presence during the pilot and Wave 1. Works directly with the customer's maintenance team to validate alerts, tune models, and build trust in the platform's outputs. This role requires someone who can explain machine learning predictions in maintenance engineering language, not data science jargon.

Requirements

5+ years in applied data science or machine learning — predictive maintenance, anomaly detection, or industrial process optimization
Strong understanding of time-series analysis, classification models (XGBoost, LightGBM), and calibration techniques
Ability to explain model behavior to non-technical maintenance engineers — "the model flagged this because superheat is trending 3 degrees above normal for this load condition" not "the feature importance vector shows..."
Comfortable working on-site at a data center or industrial facility

Nice To Haves

Experience with HVAC, mechanical, or electrical systems is a strong advantage — someone who knows what a chiller does, what COP means, why approach temperature matters

Responsibilities

Conduct the BMS protocol audit alongside the customer's controls engineer — identify available tags, data quality, polling rates, historian configuration
Extract and analyze the 30-day historian pull — identify operating modes, load patterns, seasonal variations, and data gaps
Build the baseline operating profile for each monitored equipment unit
Review the customer's maintenance logs — identify the 3–5 known fault events that will serve as ground truth validation
Map the customer's equipment taxonomy to NEIO's equipment family catalog — confirm coverage, flag gaps
Monitor live anomaly scores and triage the first alerts — determine true positive vs. false positive vs. ambiguous
Facilitate the alert review sessions with the site maintenance lead — present alerts in context, gather feedback, document site-specific operating knowledge that affects interpretation
Tune the physics constraint gate to the site's operating ranges — adjust thermodynamic bounds, load-dependent thresholds, equipment-specific parameters
Calibrate fault classification confidence thresholds — Platt scaling, ECE measurement, reliability diagram review
Validate time-to-failure estimates against known maintenance history
Document all tuning decisions and rationale — this becomes the playbook for fleet rollout
Travel to Wave 1 sites for the 1-week discovery phase at each site
Conduct on-site tag mapping review with each site's controls engineer
Facilitate the alert review session at each Wave 1 site
Train the offshore Data Scientist on site-specific tuning decisions so they can handle Wave 2 remotely