Machine Learning Engineer

Taskly Anything

United States

Remote

USD 130,000 - 190,000

Full time

14 days+

A complete application in a minute

Tailored resume and cover letter, ready to send.

Get past ATS filters

Job summary

Taskly Technologies Inc. is hiring a Machine Learning Engineer to own end-to-end ranking and recommendations in a fully remote, international team.

You will design models that decide which Taskers see new tasks, rank offers by price and fit, and test changes via A/B experiments. You will advance search relevance with keyword and vector search (pgvector) and build robust fraud signals, with a focus on cold-start handling for new Taskers and cities.

Qualifications

  • 4+ years of applied machine learning with ranking, recommendation, search or fraud systems built and run in production.
  • Strong Python and SQL, fluency in PyTorch, scikit-learn or gradient boosting libraries like XGBoost or LightGBM.
  • Experience with learning-to-rank and metrics such as NDCG, MRR and precision at k.
  • Capability to serve your model behind an API in Python or TypeScript and monitor its live behavior.
  • Ability to explain ranking decisions to non-specialists without jargon.

Responsibilities

  • Design and ship ranking model to decide which Taskers see new tasks.
  • Rank offers using price, reviews, response time and task fit; test changes with A/B experiments.
  • Improve search relevance with keyword and vector search over Postgres (pgvector).
  • Build fraud and abuse signals and enable human review for each case.
  • Define offline metrics and online experiments before deploying models.
  • Handle cold-start for new Taskers and cities using priors from similar areas.
  • Deploy notebook models to production with feature pipelines, API serving and drift monitoring.

Skills

Python
SQL
PyTorch
scikit-learn
XGBoost
LightGBM
Learning to Rank
A/B Testing
pgvector
PostgreSQL
MLOps

Tools

PostgreSQL
pgvector

Job description

Taskly is hiring a Machine Learning Engineer (also called an ML Engineer, or a Data Scientist on ranking) to decide who sees what in a two-sided marketplace. When a poster describes a task, the right Taskers should hear about it first. The poster should then see the offers most likely to end in a job done well.

You will own ranking end to end, working with the founder and the engineers who build the Tasker feed and the offer screens. The team is small and fully remote, so you will choose your own tools and defend them in writing. Within six months, we want a ranking model running behind an experiment. Its metric should show whether posters hire faster and Taskers get work that fits them.

What you’ll do
  • Design and ship the model that decides which Taskers see a new task. Balance skill fit, distance, response history and fairness to newer Taskers with no track record yet.
  • Rank the offers a poster compares, using price, reviews, response time and fit to the task. Test each change to the ordering with an A/B experiment.
  • Improve search relevance for tasks and Taskers by combining keyword and vector search over Postgres with pgvector, and measure it on real queries.
  • Build fraud and abuse signals, such as duplicate accounts, attempts to move a conversation off Taskly and unusual offer patterns. A human reviewer should be able to understand and act on each one.
  • Define offline metrics and online experiments before you build, and write down plainly when a model change did not help.
  • Handle cold start for new Taskers and new cities, where there is little history. Use priors from similar categories or neighborhoods.
  • Take models from notebook to production: feature pipelines, serving behind an API, monitoring for drift and a safe way to roll back.
What you’ll need
  • 4+ years of applied machine learning, with at least one ranking, recommendation, search or fraud system that you built and ran in production.
  • Strong Python and SQL, and fluency in PyTorch, scikit-learn or gradient boosting libraries such as XGBoost or LightGBM.
  • Practical experience with learning-to-rank and the metrics that go with it: NDCG, MRR and precision at k.
  • You have designed and read A/B tests, and you know how small samples, novelty effects and interference between users mislead.
  • You can serve your own model behind an API, in Python or TypeScript, and watch its behavior once it is live.
  • You can explain why a model ranked something first to a non-specialist in plain language, without hiding behind jargon.
Nice to have
  • Experience at a marketplace (Airbnb, Uber, Urban Company, Swiggy, Airtasker, TaskRabbit, Upwork) or another two-sided product.
  • Cold-start methods such as bandits, exploration and priors borrowed from similar cities or categories.
  • Embeddings and vector search with pgvector, including hybrid keyword and semantic ranking.
  • Work on fairness or explainability in ranking, so new Taskers get a real chance at their first job.
Skills and tools:
  • Python, SQL, PyTorch, scikit-learn, XGBoost, Learning to Rank, Recommender Systems, Search Relevance, A/B Testing, Feature Engineering, pgvector, MLOps, PostgreSQL.

Where you’ll work: Fully remote. Open to people who live and can legally work in India, the United States, Canada or the United Arab Emirates. You work from your own country, so there is no relocation and no visa to arrange.

Taskly Anything is a marketplace for getting things done, online or in person. Someone posts a task in their own words (typed, spoken or as a photo) and people who have been reviewed and approved send offers at a firm price. They pick one, and it gets done: a TV mounted, a website built, a parent taught to use their phone. We’re live in Canada and India and opening to more countries, built by a small team that works remotely across India, the US, Canada and the UAE.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

AI Engineer
AI Engineer

Taskly Anything • United States

Remote
USD 120,000 - 180,000
Analytics Engineer
Analytics Engineer

Taskly Anything • United States

Remote
USD 110,000 - 170,000
Data Analyst
Data Analyst

Taskly Anything • United States

Remote
USD 65,000 - 100,000
AI Product Manager
AI Product Manager

Taskly Anything • United States

Remote
USD 120,000 - 190,000
Remote ML Engineer — Ranking & Recommender Systems
Remote ML Engineer — Ranking & Recommender Systems

Taskly Anything • United States

Remote
USD 130,000 - 190,000
Staff Machine Learning Engineer - Ranking System
Staff Machine Learning Engineer - Ranking System

Taskrabbit • New York (NY)

Hybrid
USD 175,000 - 230,000
Health insurance
401k match
Generous PTO
Performance Marketing Manager
Performance Marketing Manager

Taskly Anything • United States

Remote
USD 90,000 - 140,000
Product Manager
Product Manager

Taskly Anything • United States

Remote
USD 120,000 - 160,000
Data Engineer
Data Engineer

Taskly Anything • United States

Remote
USD 110,000 - 150,000
Full Stack Developer
Full Stack Developer

Taskly Anything • United States

Remote
USD 110,000 - 170,000