Computer & Information Sciences
Data Science Challenges
Real data-science projects and challenges on Ewance — clean messy datasets, build and evaluate models, and turn raw data into decisions the way a working data scientist does. Solve them to build a portfolio of verified, recruiter-checkable proof you can do the work — not just describe it.
Recommended challenges
- CodeBeginnerNew
Calibrate a Multi-Camera Rig for Warehouse Robotics
You will design and prototype a calibration workflow using a printed ChArUco board (a chessboard with embedded ArUco markers). You receive a sample dataset of 200 raw frames per…
- Camera Calibration
- Multi View Geometry
- Opencv
3D Vision and Multi-View Geometry - DesignIntermediateNew
Co-Design a Trust Layer for an Enterprise RAG Assistant
You will plan and run a 5-day remote co-design study with eight pilot users (a mix of plant operators and middle managers). Sessions 1-2: discover where trust breaks down. Sessi…
- Co Design
- User Research
- Trust And Transparency
Human-Computer Interaction for AI Systems - DesignBeginnerNew
Redesign an AI Chat Sidebar for an Edtech Tutor
You receive a short Loom walkthrough of the live product, a CSV of 5,000 anonymized teacher-flagged sessions, and three teacher interview transcripts. Audit the existing sidebar…
- User Centred Design
- Heuristic Evaluation
- Interaction Design
Human-Computer Interaction for AI Systems - CodeIntermediateNew
Teach a Warehouse Cobot from Operator Demonstrations
You receive a simulated UR5e cobot in PyBullet, plus 12 example demonstrations of two kitting sequences. Implement Dynamic Movement Primitives (DMPs — a classic LfD technique th…
- Learning From Demonstration
- Dynamic Movement Primitives
- Human Robot Interaction
Human-Robot Interaction Practice your coursework on real scenarios.
Every challenge is shaped from real-world context — not generic exercises. The work mirrors what your degree prepares you for.
Why Ewance
- ResearchIntermediateNew
Evaluate a Knowledge-Graph-Augmented Recommender
You receive permission to use the public MovieLens 1M dataset plus a derived item-KG (movie -> genre, director, decade) built from Wikidata. Train two recommenders: a matrix-fac…
- Knowledge Graph Embeddings
- Recommender Systems
- Benchmarking
Knowledge Graphs and Semantic Web - ResearchSeniorNew
Investigate Scaling Trends on a Small Open Benchmark
You will train 4 transformer language models (10M, 50M, 200M, 600M parameters) on a public pretraining corpus (e.g., a small subset of FineWeb or OpenWebText) under identical op…
- Scaling Laws
- Transformer Pretraining
- Compute Optimal Training
Large Language Models - CodeBeginnerNew
Simulated Annealing for Shift Scheduling at a Hospital
You receive 6 months of anonymized shift demand data, the nurse roster (skills, certifications, contracted hours), and the labor-law hard constraints. Encode the schedule as a 7…
- Simulated Annealing
- Metaheuristics
- Constraint Handling
Evolutionary Computation and Metaheuristic Search - CodeBeginnerNew
Behavior Cloning for a Pick-and-Place Manipulator
You receive 200 human teleoperated demonstrations (state + action trajectories) of picking 8 small electronic components from a tray and placing them at marked locations in a ro…
- Behavior Cloning
- Imitation Learning
- Manipulation
Robot Learning - Browse challenges
Explore role
Marketing Analyst
Plan and measure campaigns that grow the business. Funnel analytics, attribution, segmentation, and the rigorous measurement that lets marketing defend its budget at the leadership table.
- DesignSeniorNew
Design a Distributed Training Job for a 13B-Parameter Model
Decide whether to use Fully Sharded Data Parallel (FSDP), Tensor Parallelism, Pipeline Parallelism, or a hybrid; justify against the 13B-param + 32-H100 setup. Calculate memory …
- Distributed Training
- Fsdp
- Pytorch Or Tensorflow
Machine Learning Systems - AnalysisBeginnerNew
Interpretable-by-Design GAM for an Insurer's Claims Triage
You receive an anonymized claims dataset (around 60,000 claims, target: log reserve), a feature schema (22 features), and an existing LightGBM baseline (held-out R^2 of 0.78). T…
- Generalized Additive Models
- Ebm
- Interpretability
Explainable and Interpretable AI - CodeIntermediateNew
Fine-Tune a Transformer for Customer-Support Triage at an Enterprise AI Vendor
You receive 240,000 labeled support tickets across 14 queues, with English, Bahasa Indonesia, and Tagalog. Fine-tune a multilingual transformer encoder (XLM-RoBERTa-base is a st…
- Hugging Face Transformers
- Fine Tuning
- Multilingual NLP
Deep Learning - CodeIntermediateNew
Build an Audio-Visual Speaker Diarization Pipeline
Build the pipeline: face detection + active-speaker detection on video, voice-activity detection + speaker embeddings on audio, then a fusion step that ties tracks to detected f…
- Audio Visual Fusion
- Speaker Diarization
- Active Speaker Detection
Multimodal Machine Learning Build a verifiable portfolio.
Submissions become evidence. Reviewers with shipping experience score against a rubric; the result becomes a credential anyone can verify.
Why Ewance
- AnalysisIntermediateNew
Compare ML Compiler Stacks on a Vision Backbone
Take a frozen ResNet-50 (or similar) in ONNX. Compile and benchmark it via TensorRT on Jetson + GPU, ONNX Runtime on all three, OpenVINO on x86 CPU, and IREE on ARM if time allo…
- Ml Compilers
- Tensorrt
- Onnx Optimization
Machine Learning Systems - CodeIntermediateNew
Extract Structured Lease Terms for a Commercial Real-Estate Platform
You receive 500 anonymized lease PDFs and a labelled gold set of 150 leases with the 14 fields filled in. Build a pipeline that does (1) layout-aware PDF parsing (Unstructured, …
- Information Extraction
- Pdf Parsing
- Named Entity Recognition
Linguistic Engineering and Language Technologies - CodeIntermediateNew
DPO Fine-Tune for a Domain-Specific Writing Assistant
You receive a base instruction-tuned model checkpoint plus 2,500 preference pairs from editorial reviews (each pair: two grant-application paragraphs, the editor-preferred winne…
- Dpo
- Preference Learning
- Model Finetuning
Machine Learning from Human Preferences (RLHF and Alignment) - CodeIntermediateNew
Build a Cross-Lingual Retrieval-Augmented QA System
Index around 5,000 internal-knowledge docs across the three languages using a multilingual embedding model (e.g., multilingual-e5 or BGE-M3). Build the retrieval-then-answer pip…
- RAG Architectures
- Cross Lingual Retrieval
- Multilingual Embeddings
Neural Networks for NLP - ResearchSeniorNew
Quantify Sim-to-Real Gap for a Warehouse Manipulation Policy
You receive a trained pick-and-place policy (PyTorch), the simulation env (Isaac Lab), and access to a real-arm rig (or recorded teleop episodes if hardware is unavailable). Def…
- Sim To Real
- Manipulation
- Experimental Design
Robot Perception and Autonomy - DesignBeginnerNew
Design the Onboarding for a Consumer-AI Copilot App
You receive: current funnel analytics (1M trial sign-ups over the last 90 days, current activation rate, current D1/D7/D30 retention), a competitor teardown deck, and the founde…
- Product Design
- User Onboarding
- A/B Testing & Experimentation
AI for Business and AI Product Management - CodeIntermediateNew
Build a Neural Surrogate for Computational Fluid Dynamics in HVAC Design
Use a published CFD dataset (e.g., AirfRANS or a small in-house dataset if available) of around 1,000 steady-state airflow simulations on 2D building zones. Train a Fourier Neur…
- Neural Operators
- Surrogate Modeling
- Computational Fluid Dynamics
AI for Science and Engineering - CodeBeginnerNew
Fuzzy-Logic Controller for a Sustainable-Greenhouse Operator
You receive a year of 15-minute climate logs (inside/outside temperature, humidity, light, CO2), the current rule-based controller, and the head grower's qualitative description…
- Fuzzy Logic
- Mamdani Inference
- Rule Based Systems
Fuzzy Logic, Knowledge Representation, and Symbolic Reasoning - CodeIntermediateNew
Few-Shot Defect Classifier for a Fast-Onboarding Industrial AI Vendor
You receive a multi-customer defect dataset (8 historical customers, 4-6 defect classes each). Treat 6 customers as the meta-training set and 2 as the held-out 'new customer' sc…
- Meta Learning
- Few Shot Learning
- Prototypical Networks
Meta-Learning, Transfer Learning, and Multi-Task Learning - AnalysisIntermediateNew
A/B Testing for a 40-Person SaaS Scale-up Moving to Enterprise
You are a data analyst at TaskFlow. You are given the raw A/B test data (visitor logs, conversions, and downstream sales data). Your task is to perform a rigorous analysis: chec…
- A/B Testing
- Statistical Analysis
- Bayesian Methods
Data Science for Business - ResearchIntermediateNew
Multi-Task Learning for a Healthtech Triage Model
You receive 40,000 anonymized de-identified intake-form records with two labels: urgency tier (4 classes) and routed sub-specialty (12 classes). Train (1) two independent classi…
- Multi Task Learning
- Transfer Learning
- Hugging Face Transformers
Meta-Learning, Transfer Learning, and Multi-Task Learning - CodeIntermediateNew
Multi-Turn Dialogue Manager for a Banking Assistant
You receive a transcript dataset of 200 conversations (human-tagged with intent, slot values, and required outcome), a list of 8 supported intents, and tool stubs for 3 backend …
- Dialogue Management
- Intent Classification
- Slot Filling
Question Answering and Conversational Systems
How it works
From brief to credential, in six steps.
Step 01
Browse challenges aligned to your studies.
Step 02
Accept the one that fits your goals.
Step 03
Work through it with AI Copilot guidance.
Step 04
Submit for structured evaluation.
Step 05
Earn a verified credential.
Step 06
Add it to LinkedIn with one click.
Industry teams behind a decade of practitioner briefs
Hiring from this pool?
Sponsor a challenge and meet candidates through actual work.
Industry teams can shape briefs around the skills they hire for, then evaluate students on rubric-scored deliverables — not resumes.