Machine Learning Engineer, Action Recognition

NestAI; Turku; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Occupational healthcare (Mehiläinen)

Epassi Flex benefit

Lunch benefit

Regular team lunches

Mac or PC choice

Holiday allowance

Phone benefit

Ongoing benefits expansion

Job summary

NestAI is expanding its Edge AI Chapter to advance the ATR pipeline for unmanned and command systems. You will lead action recognition work, shape how actions are modeled across objects, and guide junior engineers while delivering production-ready video analysis components.

The role blends research and hands-on development with mentoring, across a senior/lead level, as part of NestAI's Pegasus project and its battle-tested ATR reporting flow.

Qualifications

• Hands-on video analytics experience.
• Proven hands-on experience detecting and classifying actions from video streams (temporal/spatio-temporal).
• Ability to design and build a taxonomy of actions across object classes.
• Understanding of how action recognition fits into the ATR pipeline (detector, tracker, classifier).
• Experience with streaming frameworks like GStreamer/DeepStream is a plus.

Responsibilities

• Develop action recognition models using temporal information from tracked objects.
• Build and maintain a taxonomy of actions tailored to different object classes.
• Collaborate with detector and tracking teams to feed actions into the ATR pipeline.
• Mentor and guide junior team members.

Skills

Video analytics

Action recognition

Taxonomy design

Production experience

Mentoring

Tools

GStreamer

DeepStream

DLStreamer

3D vision

Job description

Company

NestAI is a European AI lab for defence delivering the adaptive operating system for modern battlefield operations. We develop adaptive intelligence for unmanned and command systems: AI that continuously learns from operational data and adapts to changing conditions.

NestAI is a European AI lab for defence delivering the adaptive operating system for modern battlefield operations. We develop adaptive intelligence for unmanned and command systems: AI that continuously learns from operational data and adapts to changing conditions. NestOS is our open, modular and interoperable platform that enables this at scale and keeps capability evolution under sovereign European control. Founded in 2025 by Peter Sarlin, whose previous company Silo AI became Europe's largest AI acquisition, NestAI brings together over 200 engineers and scientists and partners with leading European defence forces and industry.

Role Description

You’ll join NestAI’s Edge AI Chapter, working on our Automatic Target Recognition (ATR) pipeline for unmanned and command systems, as part of the Pegasus project. This role is focused on action recognition: detecting and classifying what objects are doing from video streams, and building the taxonomy of actions that different object classes can perform.

This is a senior/lead role: you’ll help shape how action recognition fits into the wider ATR pipeline, and guide junior engineers on the team.

Day-to-Day Responsibilities

• Developing action recognition models that leverage temporal information from tracked objects.
• Building and maintaining a taxonomy of actions tailored to different object classes (e.g. humans vs. vehicles).
• Collaborating with the detector and tracking teams to feed classified actions into NestAI's ATR / S.A.L.U.T.E. reporting pipeline.
• Mentoring and helping guide junior team members.

Qualifications

• Previous experience in an industry context, with a track record of taking systems into production.
• Hands‑on video analytics experience.
• Proven hands‑on experience detecting and classifying actions from video streams (temporal/spatio-temporal action recognition).
• Ability to design and build a taxonomy of actions across different object classes.
• Understanding of how action recognition fits into the broader Automatic Target Recognition (ATR) pipeline (detector, tracker, classifier).
• Experience with GStreamer, DeepStream (NVIDIA) or DLStreamer (Intel) is a plus.
• Background in 3D vision is a plus.
• Experience with temporal/sequence modeling architectures for video (e.g. 3D CNNs, video transformers) is a plus.
• Background in building action taxonomies or ontologies for computer vision systems is a plus.

How you think, learn, and take ownership matters more to us than ticking every box.

Why join NestAI

At NestAI, you’ll build meaningful, real-world technology as part of a world‑class team of engineers, scientists, and experienced professionals. You’ll work on systems where reliability genuinely matters, contribute to Europe’s security and resilience, and solve problems that demand depth, trust, and craftsmanship.

If you want to build with purpose - and with leading experts in the field - you’ve come to the right place.

What we offer

We offer a growing set of practical benefits to support your work and wellbeing, day to day:

• Occupational healthcare (currently provided by Mehiläinen)
• Epassi Flex benefit (sports, culture, commuting, wellbeing)
• Lunch benefit (25% tax‑free coverage)
• Regular team lunches and all‑hands events
• Your choice of Mac or PC
• Holiday allowance (50% of holiday salary)
• Phone benefit according to company policy (subscription and device)
• We are continuously developing and expanding our benefits as we grow.

Practicalities

Employment is subject to applicable security screening (including SUPO, where required).

NestAI is an equal opportunity employer. We consider all applicants based on their skills, experience, and potential.