{"id":27198619,"url":"https://github.com/valeqm/latam-ml-challenge","last_synced_at":"2025-04-09T20:53:09.429Z","repository":{"id":192158413,"uuid":"686163659","full_name":"valeqm/Latam-ML-challenge","owner":"valeqm","description":"Implementing a machine learning model to predict flight delays at SCL airport, deploying it as a FastAPI service, and operationalizing it on Google Cloud with proper CI/CD pipelines.","archived":false,"fork":false,"pushed_at":"2024-01-03T15:56:39.000Z","size":1751,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-09T20:53:02.595Z","etag":null,"topics":["api","fastapi","gcp","google-cloud-platform","machine-learning","python","unit-testing","unittest"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/valeqm.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-09-01T23:03:49.000Z","updated_at":"2025-01-23T05:55:58.000Z","dependencies_parsed_at":"2024-01-03T17:03:21.250Z","dependency_job_id":null,"html_url":"https://github.com/valeqm/Latam-ML-challenge","commit_stats":null,"previous_names":["valeqm/latam-challenge","valeqm/latam-ml-challenge"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/valeqm%2FLatam-ML-challenge","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/valeqm%2FLatam-ML-challenge/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/valeqm%2FLatam-ML-challenge/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/valeqm%2FLatam-ML-challenge/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/valeqm","download_url":"https://codeload.github.com/valeqm/Latam-ML-challenge/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248111976,"owners_count":21049577,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["api","fastapi","gcp","google-cloud-platform","machine-learning","python","unit-testing","unittest"],"created_at":"2025-04-09T20:53:05.567Z","updated_at":"2025-04-09T20:53:09.407Z","avatar_url":"https://github.com/valeqm.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Software Engineer (ML \u0026 LLMs) Challenge\n\n## Overview\n\nWelcome to the **Software Engineer (ML \u0026 LLMs)** Application Challenge. In this, you will have the opportunity to get closer to a part of the reality of the role, and demonstrate your skills and knowledge in machine learning and cloud.\n\n## Problem\n\nA jupyter notebook (`exploration.ipynb`) has been provided with the work of a Data Scientist (from now on, the DS). The DS, trained a model to predict the probability of **delay** for a flight taking off or landing at SCL airport. The model was trained with public and real data, below we provide you with the description of the dataset:\n\n|Column|Description|\n|-----|-----------|\n|`Fecha-I`|Scheduled date and time of the flight.|\n|`Vlo-I`|Scheduled flight number.|\n|`Ori-I`|Programmed origin city code.|\n|`Des-I`|Programmed destination city code.|\n|`Emp-I`|Scheduled flight airline code.|\n|`Fecha-O`|Date and time of flight operation.|\n|`Vlo-O`|Flight operation number of the flight.|\n|`Ori-O`|Operation origin city code.|\n|`Des-O`|Operation destination city code.|\n|`Emp-O`|Airline code of the operated flight.|\n|`DIA`|Day of the month of flight operation.|\n|`MES`|Number of the month of operation of the flight.|\n|`AÑO`|Year of flight operation.|\n|`DIANOM`|Day of the week of flight operation.|\n|`TIPOVUELO`|Type of flight, I =International, N =National.|\n|`OPERA`|Name of the airline that operates.|\n|`SIGLAORI`|Name city of origin.|\n|`SIGLADES`|Destination city name.|\n\nIn addition, the DS considered relevant the creation of the following columns:\n\n|Column|Description|\n|-----|-----------|\n|`high_season`|1 if `Date-I` is between Dec-15 and Mar-3, or Jul-15 and Jul-31, or Sep-11 and Sep-30, 0 otherwise.|\n|`min_diff`|difference in minutes between `Date-O` and `Date-I`|\n|`period_day`|morning (between 5:00 and 11:59), afternoon (between 12:00 and 18:59) and night (between 19:00 and 4:59), based on `Date-I`.|\n|`delay`|1 if `min_diff` \u003e 15, 0 if not.|\n\n## Challenge\n\n### Instructions\n\n1. Create a repository in **github** and copy all the challenge content into it. Remember that the repository must be **public**.\n\n2. Use the **main** branch for any official release that we should review. It is highly recommended to use [GitFlow](https://www.atlassian.com/git/tutorials/comparing-workflows/gitflow-workflow) development practices. **NOTE: do not delete your development branches.**\n   \n3. Please, do not change the structure of the challenge (names of folders and files).\n   \n4. All the documentation and explanations that you have to give us must go in the `challenge.md` file inside `docs` folder.\n\n5. To send your challenge, you must do a `POST` request to:\n    `https://advana-challenge-check-api-cr-k4hdbggvoq-uc.a.run.app/software-engineer`\n    This is an example of the `body` you must send:\n    ```json\n    {\n      \"name\": \"Juan Perez\",\n      \"mail\": \"juan.perez@example.com\",\n      \"github_url\": \"https://github.com/juanperez/latam-challenge.git\",\n      \"api_url\": \"https://juan-perez.api\"\n    }\n    ```\n    ##### ***PLEASE, SEND THE REQUEST ONCE.***\n\n    If your request was successful, you will receive this message:\n    ```json\n    {\n      \"status\": \"OK\",\n      \"detail\": \"your request was received\"\n    }\n    ```\n\n\n***NOTE: We recommend to send the challenge even if you didn't manage to finish all the parts.***\n\n### Context:\n\nWe need to operationalize the data science work for the airport team. For this, we have decided to enable an `API` in which they can consult the delay prediction of a flight.\n\n*We recommend reading the entire challenge (all its parts) before you start developing.*\n\n### Part I\n\nIn order to operationalize the model, transcribe the `.ipynb` file into the `model.py` file:\n\n- If you find any bug, fix it.\n- The DS proposed a few models in the end. Choose the best model at your discretion, argue why. **It is not necessary to make improvements to the model.**\n- Apply all the good programming practices that you consider necessary in this item.\n- The model should pass the tests by running `make model-test`.\n\n\u003e **Note:**\n\u003e - **You cannot** remove or change the name or arguments of **provided** methods.\n\u003e - **You can** change/complete the implementation of the provided methods.\n\u003e - **You can** create the extra classes and methods you deem necessary.\n\n### Part II\n\nDeploy the model in an `API` with `FastAPI` using the `api.py` file.\n\n- The `API` should pass the tests by running `make api-test`.\n\n\u003e **Note:** \n\u003e - **You cannot** use other framework.\n\n### Part III\n\nDeploy the `API` in your favorite cloud provider (we recomend to use GCP).\n\n- Put the `API`'s url in the `Makefile` (`line 26`).\n- The `API` should pass the tests by running `make stress-test`.\n\n\u003e **Note:** \n\u003e - **It is important that the API is deployed until we review the tests.**\n\n### Part IV\n\nWe are looking for a proper `CI/CD` implementation for this development.\n\n- Create a new folder called `.github` and copy the `workflows` folder that we provided inside it.\n- Complete both `ci.yml` and `cd.yml`(consider what you did in the previous parts).","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvaleqm%2Flatam-ml-challenge","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvaleqm%2Flatam-ml-challenge","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvaleqm%2Flatam-ml-challenge/lists"}