About the Role

We're looking for a hungry, hands-on AI Engineer to join our data science team. You'll do the work directly, labeling and reviewing classification data and running QC, but you won't just execute. You'll bring an engineer's mindset to it: when a task is repetitive, you script it; when quality is hard to measure, you build a way to measure it. You'll use Python, SQL, and agentic development tools to make annotation and QC faster, more consistent, and more scalable.

This is a rapidly evolving role, and we expect you to context switch comfortably as priorities shift. You'll work shoulder-to-shoulder with data scientists and ML engineers, people who think about data the way you do, and the labels and quality signals you produce feed directly into the models that protect real customers' most sensitive data. The work is high-impact and the data is messy; a big part of the job is learning, through the work itself, what it takes to make it usable.

This is a hybrid role requiring 3+ days in-office in New York City.
Who Should Apply

We're looking for someone with programming ability, dependability, and the drive to learn on the job. Recent grads are welcome, and CS and STEM backgrounds are a great fit. What matters most is that you can think critically, you're excited to work through messy data, you can context switch as priorities change, and you want to grow fast in a fast-moving environment.
What You'll Do

• Do hands-on data annotation and quality control (labeling, reviewing, and correcting classification outputs) as a core member of the data science pipeline.
• Take ownership of improving and scaling the process: find the bottlenecks, repetitive steps, and sources of error, and fix them with Python, SQL, and agentic workflows.
• Build and run quality control checks that catch labeling errors, measure inter-annotator agreement, and surface systematic issues before they reach production.
• Work closely with data scientists and ML engineers to close the loop between real-world performance and model improvement.
• Context switch across labeling, quality analysis, scripting, and process work as priorities evolve.
• Document QC processes and annotation guidelines to support team scaling and onboarding.

About You

• Solid programming ability, with hands-on Python experience and a willingness to dig into scripts, SQL, and data wrangling.
• Comfortable using agentic development tools, or eager to ramp up on them fast.
• A quality-first mindset. You notice when something is off in the data and won't let it slide.
• Dependable and adaptable. Teammates can count on you, and you stay effective as priorities shift.
• Energized by messy, real-world data and by working alongside other data-minded people.
• Hungry, self-directed, and ready to grow with Teleskope as we scale.

Nice to Have

• Familiarity with feedback loops in ML systems and how label quality connects to model performance.
• Experience with annotation platforms (Label Studio, Prodigy, Scale, or custom-built systems).
• Familiarity with active learning or online learning approaches.
• Experience with SQL and building lightweight dashboards to track quality metrics.
• Background in NLP or text classification workflows.

What You'll Get

• A seat alongside data scientists and ML engineers, data-minded people to learn from every day.
• Work that visibly matters. Your labels feed the models that protect real customers' most sensitive data.
• Ownership of the annotation and quality processes that determine classification accuracy across the platform.
• Room to grow fast, with real ownership from day one as Teleskope scales.
• A beautiful, well-stocked office in NYC's Financial District.
• Flexible vacation and work-from-home days.
• Competitive salary and meaningful equity.
• Health, vision, dental, 401k, and more benefits, heavily subsidized by Teleskope.

What We Value

At Teleskope, we value builders who care about the details. This role is for someone who sees data quality not as a support function but as a force multiplier, and who takes pride in making the people around them more effective. We look for dependable teammates who ship iteratively, take ownership, and understand that great ML starts with great data.