Control Theory Project-  CPA Optimization 

Title:

Expert consultation for a software PID Controller Implementation and Tuning

Background:

In a nutshell, this is Taboola’s main business model:

• Taboola places its widgets on publishers’ (e.g. nbcnews.com, usatoday.com ) pages

• On each widget, we recommend content from multiple advertisers

• Every time someone clicks on a piece of content, they’re directed to the advertiser’s page, and Taboola gets a referral fee called CPC (Cost Per Click)

• The income generated is shared with the publisher, measured at RPM (Revenue Per Mille impressions) based. Mille stands for a thousand.

• Taboola chooses which item to show to a user according to an algorithm that ranks items by their expected revenue [eCTR * CPC = eRPM]. On each impression the item with the highest expected revenue is shown:

• eCTR - Estimated Click Through Rate (Clicks/Impressions)  is predicted by our algorithm

• CPC - Cost Per Click - given as an input by the advertiser set on the campaign level

• While advertisers pay per click, their marketing goals are usually defined by cost per action (CPA) and scale of actions. Action can be - purchase, lead generation etc. This means that one of the clients’ challenges is to generate Actions at scale while keeping profitability (i.e. CPA that is lower than the worth of actions) which can be achieved by adjusting the CPC of the campaign.

Overview:

Taboola’s advertisers invest a lot of time in manual optimization of their campaigns, trying to achieve the optimal balance of CPA goal and scale of actions while tuning the CPC on the campaign level and on some sub-campaign levels taboola offers. Setting the CPC too high means the advertiser bids will be very competitive but the price might be too high resulting in too high CPA. Setting the CPC too low means the advertiser bids might not be competitive and might not get enough clicks/actions or might not ‘win’ the valuable user clicks.

Our agenda is to automate this manual process of tuning the CPC by utilizing a feedback loop which will be able to optimize the Campaign-CPA towards a given CPA goal. 

We want to build a feedback loop with two (and possibly more) independent inputs that can influence on the campaign level CPA which will be evaluated and compared to the client’s input Target CPA

In addition, since there are multiple inputs we control, we believe there will be multiple points that satisfy the CPA goal. Within the points that satisfy the criteria of CPA <= Target CPA, we would like to be able to optimize for additional metrics - for example, number of actions (aks conversions scale) or maximum profit,  etc. we can use such inputs as additional indication to tune and impact the feedback loop and the campaign CPA accordingly.

Project scope:

We are looking for an expert in multivariable control systems and optimized control systems. The project will comprise an expert consultation for the tuning process, plus joint practical work with the team that is developing the controller. The deliverable will comprise several tuned instances of the controller, on choice placements (TBD) - and a simulation on real data - which will serve to prove the proposed concept for tuning the controller and for cleaning the noise. The result will be a well documented tuning procedure and noise filtering method to allow automation of the procedure by the team towards large-scale deployment.

The project has strong potential for follow-on projects based on the success of the initial implementation.  

Your proposal:

Your proposal should emphasize your background in control theory and practical implementations of control systems in software with very large scale data, as well as your background in Ad-technology or performance advertising;

You should also outline which algorithms and approach you intend to use, and how you propose to measure the success of tuning and noise filtering.

We will share further information with the expert and jointly refine the approach and scope in detail with the relevant exp

Need a great, independent Linux administrator for this long-term, on-demand contract position. Must be located near the Boston area for occassional data center visits (every 3-4 months).

Must have:

•  Advanced knowledge of Linux CentOS 6, 7, 8
•  Advanced knowledge of security and network concepts: intrusion detection, switching, routing, IPTables, Linux firewalls, IPMI
•  Extensive virtualization experience. Docker, KVM, (OpenVZ is a plus)
• Datacenter co-location experience - racking, stacking, cabling

Experience with VoIP (asterisk) and AWS (Amazon Cloud) is a big plus too.

We have:
50+ physical servers (2 racks)
300 virtual servers

I'm building an Autopilot enabled wine advisor - like an e-sommelier. I have all the information I want to put into the system, but I need a little help setting it up in the right way.

I have thousands of responses written for things like food & wine pairings, region and grape varietal descriptions, wine word glossary, etc. 

1. I want to create the right framework for putting the data into the system, so I only have to do it once

2. I'm looking for a creatiove way to import all the responses/data, or at least a smart way to do the data-entry

3. I want to make sure it is set up to be useful on a variety of platforms: SMS, voice, Alexa, etc.

General:

Taboola predicts RPM of ad impressions via a Deep Neural Network (NN). These predictions are used for ranking ad items in an auction. The predicted RPM which is the output of the NN contains bias, and requires calibration to the actual mean RPM achievable per each set of environmental parameters.

A PID controller performs the task of calibrating the output of the NN. Since the clicks are sparse, the noise in the measured actual RPM is significant and requires filtering. Moreover, each supply source has its own unique attributes - so a repeatable method of tuning controller parameters at scale to varying supply is required. 

What we are looking for:

A consultation and hands-on practice - for tuning a PID controller for correcting bias in RPM predictions, and smoothing PV RPM measurement noise of said controller. 

Our goal is to create a scalable PID controller which includes (a) an RPM measurement that is inherently resistant to noise, applicable to a wide range of supply and demand characteristics: input data rates, click through rates, CPC, other demand competing in the auction - i.e. highly varied RPM ranges and measurement window durations; and (b) the controller parameters should be tunable to correct for RPM bias over many supply placements and dynamic conditions. 

A method should be devised for tuning the parameters of the controller and the noise filter for a wide range of such conditions.  

Project scope:

The project will comprise an expert consultation for the tuning process, plus joint practical work with the team that is developing the controller. The deliverable will comprise several tuned instances of the controller, on choice placements (TBD) - and a simulation on real data - which will serve to prove the proposed concept for tuning the controller and for cleaning the noise. The result will be a well documented tuning procedure and noise filtering method to allow automation of the procedure by the team towards large-scale deployment.

The project has strong potential for follow-on projects and scope enlargement based on the success of the initial cooperation.  

Your proposal:

Your proposal should emphasize your background in control theory and practical implementations of control systems in software with very large scale data, as well as your background in Ad-technology and programmatic advertising or display advertising;

You should also outline which algorithms and approach you intend to use, and how you propose to measure the success of tuning and noise filtering.

We will share further information with the expert and jointly refine the approach and scope in detail with the relevant expert.

Note: the timezone of the customer is GMT+2, so there would be an advantage in communicating if near that timezone. But this is not a mandatory requirement. 

We have a client who is hosting a large Drupal site on AWS. We need to speed up the site, pages are taking too long to load. Cloudfront would be a good start. We are looking for someone to not only identify issues but someone who can help us execute the solutions.

We are a Fortune500 financial firm seeking a GoLang developer

The developer will be building a centralized customer information portal for our firm. The portal will be developed in Golang, and will expose REST endpoints and accept data streams via Kafka.

• Golang experience is a must.
• Experience with Postgres DB, Kafka, and REACT would be a big plus

Skills required:

1. Expertise with Python, Jango and React software, in addition to Natural Language Processing libraries (SpaCy, NLTK, etc)
2. Expertise in sentiment analysis and deep learning models to make predictions (bonus if they can build predictive models in the financial markets using text modeling)
3. Expertise in leveraging APIs to extract financial and government data, news articles and filings
4. Experience with financial markets and investment preferred.

https://www.nytimes.com/2019/01/10/business/secrets-corporate-reports-apple.html

https://www.youtube.com/watch?v=g96gROyc3wE

https://www.nber.org/papers/w25084.pdf

Looking for a skilled developer in Angular native/Angular React that can build a responsive app that integrates with a MYSQL database.

Resource should be able to quickly translate functional wireframes to create pages to be reviewed and revised by the functional design team.

Technical Qualifications:
• 5-6 years of experience in software development.
• 5-6 years of Angular Native or Angular React with JavaScript/CSS/HTML
• 5-6 years of solid experience with MySQL
• Hands on experience with Application based frameworks
• Hands on experience in translating functional wireframes in to web pages
• Experience with CI/CD systems such as Jenkins
• Experience with XML/JSON and developing REST Services
• Experience with UNIX utilities &amp; UNIX Commands
• Ability to understand complex systems and solve challenging analytical problems
• Ability to leverage multiple tools and programming languages to analyze and manipulate
data sets from disparate data sources.
• Strong collaboration and communication skills within and across teams.

Currently looking for US-based consultants.

We are looking for a Data Scientist who has extensive experience with End to End ML in Azure

Objective:

• Create a small end-to-end demo of Azure's machine learning capabilities

Deliverables:

• Source code / notebooks that demonstrate how each service is used

• A short presentation of the findings

Project phases:

Phase 0: EDA 

• Go over a sample of the data set using Azure Machine Learning / Azure Machine Learning studio

• Understand the framework's capability to detect interactions between the dependent variable and independent variables

Phase 1: Data wrangling

• Generate a training set using Azure Databricks

• Down sample data 

Phase 2: Model building 

• Use the framework to build a model

• Deal with hyper parameter optimization

• Use out of the box feature engineering

• Display training metrics

Phase 3: End-to-end service

• Upload data to Azure data lake

• Create a model container using Azure Container Registry / Azure Container Instance

• Write a small Azure function to take some data lake records and run them through the container

We are currently only considering candidates based in the United States.

Looking for Jira Architect for Fortune 100 company who has hands-on experience who can craft the workflows based on business requirements.

Responsibilities:
• Work with a cross-functional team to gather requirements and inputs needed to determine
clearly defined JIRA workflows and architecture.
• Work with a cross-functional team to gather requirements and inputs needed to establish
JIRA dashboards to be used for easy high-level snapshots of project progress.
• Work with a cross-functional team to gather requirements and inputs needed to create
standardized reports that can be run and exported with ease.
• Establish a standardized Jira workspace template that will be used to house all materials
relevant to the web development processes.
• Create standardized templates for functional requirements document, technical architecture
documentation, and user stories.
• Assist with development of training documentation that can be used to roll out JIRA,
Confluence, and the requirements documents to our global organization
• Administer and maintain JIRA environments post-implementation; suggest and implement
optimizations based on user behavior.

Technical Qualifications:
• BA/BS degree
• 5+ years of relevant experience with technical writing, JIRA and Confluence
administration and implementation across an organization.
• Must have experience with reporting capabilities and functionality within JIRA.
• Established track record of developing requirements documentation to support web
development processes.
• Ability to setup security permissions and policies.
• Ability to setup new rules and triggers and new fields in a Jira workflow and
form.
• Proven ability to work with cross-functional teams, including marketing, strategy, GIS/IT,
and others.
• Experience with reporting capabilities and functionality within JIRA and familiar with
Agile environment.
• Self-motivated and able to prioritize and manage workload and meet critical project
milestones and deadlines
• Excellent interpersonal, communication, negotiation and collaboration skills, including
the ability to lead working sessions and workshops.

Currently, looking for US-based candidates.