Join Halter as a Manufacturing Data Scientist to analyze manufacturing data and improve cattle management technology.
Posted by employer 1 day ago
First seen on Joblaze 1 hour ago
Last verified on the company career page 1 hour ago
What you'll build
Must have
Nice to have
Practical constraints
AI in the day-to-day
Use AI tools to move fast in data analysis.
Not disclosed in this posting: compensation, seniority, years of experience, visa sponsorship.
Joblaze summary
The Manufacturing Data Scientist at Halter focuses on analyzing and modeling manufacturing data to identify potential issues and improve product quality. Proficiency in Python, R, or MATLAB, along with a solid understanding of applied statistics and basic electronics, is essential for this role. It is well-suited for candidates with experience in data science and a background in hardware or manufacturing environments. The position involves close collaboration with engineers and offers the opportunity to influence production processes directly.
Joblaze insights
Quick facts
From the original posting
We test every collar we build, and we have more manufacturing data than we can use. Turn it into models and answers that catch problems early, and understand the hardware well enough to know when a number is physically wrong.
Halter's solar-powered smart collars let farmers move, monitor and manage their cattle from a phone.
This role sits with our test engineers, not in a business intelligence team. You'll work on test and process data from physical electronics, and we're only scratching the surface of what modelling could do with it. You don't need to live on the bench, but you'll understand how power, voltage and current relate, go deep on the hardware when a question needs it, and help close the loop with failure analysis rather than only raising a flag. We're open on level for the right person.
Own recurring yield analysis and deep dives into failure modes.
Build models that separate real shifts from normal variation and spot risks before they reach the field.
Backtest across our data to find patterns that improve the product, the process and supplier quality.
Investigate unusual results through to a cause engineers can act on, working with them at the station when needed.
Build and maintain the dashboards and reports production and test teams use daily.
Flag and help resolve stations, fixtures or data feeds producing results that can't be trusted.
Test results from every station on our production line, for every collar built.
Station, process and field data brought together for analysis.
Python, R or MATLAB, with whichever tools suit the job, plus AI tools to move fast.
Statistical process control, control charts and measurement system analysis.
Strong data science skills in Python, R or MATLAB, including statistical modelling.
Solid applied statistics on large data sets: hypothesis testing, variation, sampling and control charts.
Experience with data from a physical product, such as hardware, electronics, automotive or manufacturing.
A working grasp of basic electronics, and the curiosity to dig into how the product works.
Clear communication of findings to engineers and managers, including through good charts.
A postgraduate degree or research background in a quantitative field.
Experience in manufacturing test, quality or failure analysis.
Measurement system analysis or gauge repeatability studies.
Experience working with software teams on data pipelines.
You'll be on site in Auckland, working closely with test, production and quality engineers, with occasional travel to our contract manufacturer. You'll have real ownership of the questions you investigate and see your findings change how the line runs.
Work that genuinely matters. Every now and again a company comes along that transforms an entire industry and leaves the world in a better place. Our team gets to be part of something truly meaningful, helping farmers improve their livelihoods, spend more time with their families, and build more sustainable operations.
Standard company text repeated across Halter's postings is omitted here.
Explore more