This is a part-time, fully remote MLOps Engineer opening, open to applicants anywhere, with pay up to $77,000 a year on a prorated part-time schedule.
The role sits inside a small AI and machine learning group where models don't stay experiments for long. They get shipped, watched, and kept alive in production. The person in this seat makes sure that happens without drama: pipelines that run on schedule, retraining that starts automatically when data drifts, and monitoring that flags problems before a customer ever notices. It's structured as part-time work, built for someone who wants real production ownership without a full-time weekly commitment.
What the work actually involves
Day-to-day, the job centers on moving models from training to deployment to monitoring without requiring a person to babysit each stage by hand. Some weeks lean toward building new pipeline capability. Other weeks lean toward tracking down why a job failed overnight. The core responsibilities include the following.
- Building and maintaining pipelines that handle model training, deployment, and ongoing monitoring in production environments
- Automating model retraining so performance decay gets caught and addressed rather than discovered weeks after the fact
- Keeping the underlying infrastructure reliable and able to scale as usage grows or shifts
- Troubleshooting pipeline and deployment failures as they arise, sometimes under time pressure
A bachelor's degree in computer science, engineering, or a closely related field is the baseline expectation for this role, and Naukri Mitra typically pairs that with real experience spanning both software engineering and machine learning workflows. This isn't a pure data science seat, and it isn't a pure infrastructure seat either. It sits between the two, connecting model development to the systems that actually run it in production.
What you'll need to bring
Thirty months of relevant experience is the baseline the team is looking for here. Comfort with containerization and cloud infrastructure isn't a nice extra; it's something this role depends on every week.
- CI/CD pipelines built and maintained in a live production setting
- Docker and Kubernetes, for both containerization and orchestration
- Working knowledge of at least one major cloud platform
- Python as a primary language for tooling and automation
- Infrastructure as code for repeatable, versioned environment setup
- Experience with MLflow, or a comparable tool for tracking and monitoring models
None of these need to be mastered at an expert level from the first week. Someone who has shipped a handful of models to production and knows what tends to break usually outperforms a candidate with only theoretical exposure to the tooling.
Schedule and pay
The $77,000 annual figure reflects the full-time-equivalent rate, prorated against actual hours worked on a part-time schedule, which puts this among the better-paying part-time remote MLOps engineer roles currently open. Because the team is distributed, most of the work happens asynchronously, with a small block of overlapping hours set aside for moments that genuinely require real-time coordination: an incident, a deployment window, a planning check-in. The scope of the role doesn't shrink just because the weekly hours do, and the technical ownership stays fully intact compared to a full-time equivalent.
Most people who end up qualified for this kind of work got there through hands-on software engineering first, machine learning exposure second, with the operational habits, monitoring, alerting, rollback discipline picked up gradually on the job rather than learned from a single course. Anyone building toward becoming a remote MLOps engineer over the next year or two is likely following a similar order. A candidate with that shape to their background is a reasonable fit for this opening rather than a stretch, even without specific prior part-time experience.
Benefits
Part-time status here doesn't mean the support around the role is an afterthought. This position includes the following.
- Prorated paid time off, so hours worked still translate into real time away from the job
- A professional development budget that can go toward cloud or MLOps certifications
- Full remote flexibility, with no fixed office and a light meeting schedule
- A stipend toward home-office equipment for anyone who needs to upgrade their setup
Together, these are meant to make the part-time arrangement genuinely sustainable rather than a compromise the candidate quietly absorbs on their own. None of these are framed as perks stacked on top of a base offer; they're treated as standard parts of how this specific position is structured, part-time hours and all.
Team and working style
The broader engineering group is small enough that decisions don't get stuck in committee and large enough that no single person understands how a given system works. Retraining jobs fail sometimes. Deployments occasionally roll back. The team treats that as routine maintenance rather than a crisis every time it happens. What matters more than whether an issue occurs is whether it gets caught early and communicated clearly to the people who need to know.
Because the role is remote and part-time, a fair amount of trust gets placed in how someone manages their own hours. Nobody is tracking keystrokes or measuring time logged in a dashboard. What gets tracked instead is simpler: whether pipelines run on schedule, whether retraining happens when it's supposed to, and whether the person covering a given window actually responds when something breaks. "Part-time" in this context refers to the number of hours, not a smaller version of the responsibility. Pipeline ownership and uptime remain the same as in a full-time seat.
How to apply
Send a resume that reflects genuine production experience with model pipelines rather than coursework or side projects alone, along with a short note on which platforms and tools you've worked with directly. If a retraining pipeline you managed ever failed silently for a stretch before anyone caught it, mention that too. Concrete, slightly uncomfortable stories from real incidents carry weight here, more than another generic bullet claiming "strong problem-solving skills" would. Applications get reviewed as they arrive, and most candidates who move forward hear back with follow-up questions within a few business days. There's no fixed closing date listed for this posting; it stays open until the role is filled, so applying earlier rather than later tends to work in a candidate's favor.
