Data Scientist

Bykea Technologies Ltd.; Karachi Division; On-site; PKR 1,800,000 - 2,400,000; Full time

Job summary

Bykea Technologies Ltd. is seeking a Data Scientist to tackle challenging problems across transport, logistics, commerce, economics, and fintech in Pakistan. You will build predictive models, work with large-scale data, and collaborate with cross‑functional teams to deploy solutions that impact products and operations.

The role emphasizes end-to-end ownership and experimentation, with opportunities to grow scope and ownership within autonomous data science units across the company.

Qualifications

• 2–5 years of professional experience in Data Science, ML, Data Engineering, BI, Data Architecture, or related technical role.
• Strong understanding of ML, DL, data mining, statistics, and optimization foundations.
• 3+ years of hands‑on experience with SQL and Python.
• Experience building and evaluating ML models using scikit-learn, Spark MLlib, or similar.
• Experience analyzing large datasets and translating insights into business impact.
• Experience implementing and reporting KPIs in data warehousing or large-scale data environments.
• Experience with distributed data processing or big-data technologies like Spark, Hive, MapReduce.
• Strong quantitative and analytical skills with ability to break down problems.
• Ability to communicate technical concepts clearly to non-technical stakeholders.
• Strong attention to detail and ability to manage multiple priorities.

Responsibilities

• Build predictive models and algorithms to solve business problems.
• Solve complex marketplace problems like allocation, matching, pricing, and routing.
• Work with large-scale data and streaming data pipelines.
• Apply ML, DL, geospatial analytics, forecasting, simulation, and NLP.
• Collaborate with Engineering, Product, Marketing, Operations, and Finance.
• Create repeatable data solutions and monitoring for value extraction from data.
• Design experiments, KPI development, and ROI analysis.
• Own problems end-to-end from data prep to deployment and monitoring.

Skills

SQL and Python

Machine learning

Deep learning

Data mining

Statistics

Optimization

Analytical thinking

Communication

Problem solving

Education

Bachelor's or Master's in Computer Science/Engineering/Math/Statistics

Tools

Python

Spark

Spark MLlib

Hive

NoSQL

PostGIS

Job description

What this role really is

Bykea's mission is to improve the movement of people and parcels in Pakistan, and data science is at the heart of how we make that movement smarter, faster, and more efficient. As a Data Scientist at Bykea, you'll work on some of the most challenging and fascinating problems across transport, logistics, commerce, economics, and fintech. You'll use machine learning, deep learning, forecasting, optimization, simulation, geospatial data, and large-scale data processing to solve real business problems and directly influence our products and operations. This role is an opportunity for someone to grow the scope and ownership. Bykea is building autonomous data science units focused on different business problems, allowing you to work deeply within an area while also contributing to broader data science capabilities across the company. This is a problem-solving role, not a reporting-only role. You'll take ambiguous business problems, turn them into analytical and modelling problems, build solutions, evaluate their impact, and work with Engineering, Product, Marketing, and Operations to put those solutions into practice. If you enjoy asking \"What can we predict, optimize, automate, or improve with data?\", this role is for you.

What you'll actually do

• Build predictive models and algorithms. Develop machine learning and statistical models to solve problems such as ETA prediction, marketplace churn, demand forecasting, customer behavior, marketplace pricing, incentives, and marketplace optimization.


• Solve complex marketplace problems. Work on intelligent allocation, passenger-driver matching, supply and demand positioning, dynamic pricing, routing, scheduling, and other problems where the right algorithm can meaningfully improve marketplace efficiency.


• Work with large-scale data. Extract insights and build models using Bykea's large datasets, distributed computing environments, data warehouses, ETL pipelines, and real-time data streams.


• Apply advanced data science techniques. Use machine learning, deep learning, geospatial data mining, forecasting, simulation, optimization, NLP, clustering, classification, and other appropriate techniques to solve real-world problems.


• Work cross-functionally. Partner with Engineering, Product, Marketing, Operations, Finance, and Customer Service teams to understand problems, define metrics, build datasets, run experiments, and turn analysis into business impact.


• Build repeatable data solutions. Create reliable datasets, modelling pipelines, monitoring systems, analytical processes, and tools that allow teams to consistently extract value from data.


• Experiment and measure impact. Contribute to A/B testing, experiment design, KPI development, business intelligence, marketing effectiveness, ROI analysis, and other measurement frameworks.


• Own problems end-to-end. Take a problem from understanding the business context and preparing the data through modelling, validation, implementation, monitoring, and iteration, seeking guidance when needed on complex technical or business decisions.

Where you could work

Our Data Science teams are being built around different business problems, so your work could sit in one or more of the following areas:

Central Platform

Work on capabilities that cut across the business, including A/B testing, forecasting, conversational AI, customer experience, driver network solutions such as facial detection, and data science solutions for teams such as Finance and Customer Service.

Ride-hailing

Build models and algorithms for passenger-driver matching, ETA prediction, driver churn, demand and supply forecasting, maps and routing, pricing, incentives, personalization, marketplace optimization, and driver and customer experience.

Risk & Fintech

Apply data science to risk management, customer and driver profiling, credit scoring, payment fraud prevention, account security, credit risk, marketplace optimization, pricing, and forecasting.

Must-haves

• 2–5 years of professional experience in Data Science, Machine Learning, Data Engineering, Business Intelligence, Data Architecture, Data Modelling, or another closely related technical role.


• Strong understanding of machine learning, deep learning, data mining, statistics, and algorithmic foundations of optimization.


• 3+ years of hands‑on experience with SQL and Python or equivalent experience demonstrating strong proficiency in both.


• Experience building and evaluating machine learning models using frameworks such as scikit-learn, Spark MLlib, or similar is a plus.


• Strong experience manipulating, querying, and analysing large datasets and translating them into meaningful business insights.


• Experience implementing and reporting on business KPIs in data warehousing or large-scale data environments.


• Experience working with distributed data processing or big-data technologies such as Spark, Hive, MapReduce, or similar is a plus.


• Strong quantitative and analytical skills, with the ability to break down complex problems and identify practical solutions.


• Working knowledge of machine learning techniques such as clustering, decision trees, classification, regression, neural networks, or similar approaches , including an understanding of when different techniques are appropriate and their trade‑offs.


• Ability to work independently, learn quickly, and communicate technical concepts clearly to non‑technical stakeholders.


• Strong attention to detail and the ability to manage multiple priorities in a fast‑moving environment.


• Be competent with using AI and automation tools effectively and efficiently to boost productivity in all relevant areas of the role, including analysis, coding, research, documentation, experimentation, and model development. Should have experience using these tools for at least a year to understand their strengths and weaknesses.


• Strong communicator and collaborative team player.

Nice‑to‑haves

• Experience building ML models at scale using real‑time or streaming data pipelines.


• Experience with NoSQL databases, PostGIS, stream processing, or distributed computing platforms.


• Experience working with geospatial data, maps, routing, location intelligence, or marketplace data.


• Experience with forecasting, optimization, simulation, recommendation systems, NLP, or experimentation.


• Experience with R or Scala in addition to Python.


• Experience working with high‑volume consumer, marketplace, logistics, mobility, e‑commerce, fintech, or similar products.


• Experience taking machine learning models from experimentation into production.

Our data science environment

• You’ll work across large-scale datasets and technologies spanning Python · SQL · scikit-learn · Spark / Spark MLlib · Hive · NoSQL · distributed computing · real‑time data pipelines · data warehousing · machine learning · deep learning · experimentation and analytics.


• The exact tools will vary by problem, we care more about your ability to choose and apply the right approach than about knowing every tool beforehand.

Experience & qualifications

• 2–5 years of relevant professional experience in Data Science, Machine Learning, Data Engineering, Business Intelligence, Data Architecture, Data Modelling, Engineering, or a related technical field.


• Bachelor's or Master's degree in Computer Science, Software Engineering, Electrical/Computer Engineering, Operations Research, Mathematics, Statistics, or a related quantitative field.


• Ph.D. candidates/graduates in relevant fields are also welcome.


• Location: Karachi, onsite.

How we'll assess

Expect a hands‑on technical assessment focused on real data science problems. We care about how you approach a problem, reason about the data, select and explain a modelling approach, evaluate trade‑offs, and communicate your solution, not just whether you can produce a technically correct answer.

What success looks like

• Machine learning and analytical solutions that solve meaningful business problems and improve measurable business metrics.


• Models and data pipelines that are reliable, scalable, and practical to use in production.


• Clear, well‑defined datasets, KPIs, experiments, and monitoring systems that help teams make better decisions.


• Strong collaboration with Engineering, Product, Marketing, Operations, Finance, and other stakeholders.


• Increasing ownership of complex data science problems from problem definition through modelling, implementation, and impact measurement.


• Continuous improvement in the techniques, processes, and technology Bykea uses to extract value from its data.