We are the first AI based corporate robo advisor focused on three areas - strategy, board and leadership assessment, and performance measurement and reporting.

We use tensor flow and publicly available data (Factset, CapIQ, Edgar, Mergent, Corporate Disclosures - EDGAR) to analyze global corporate strategies, leadership and boards and KPI’s and visualize the results via TIBCO. 

Expertise Needed:

A Data Visualization Expert that understands TIBCO, Tableau, MS Power BI and Google Data Studio that can help us improve our current TIBCO visualization and help us migrate our data vis to another, more appropriate, and web/mobile centric environment. See more details on the current Beta product enclosed.

Key skills

• Tibco Spotfire
• Data Visualization
• Additional desired competencies - data science and data architecture.

Data sources: Factset, CapIQ, Edgar, Mergent, and many nore.

Current technology stack: Google Compute, Tensor Flow, Tiibco and AWS. 

Deliverable: Data Visualization -

Phase 1: Improve our TIBCO environment,

Phase 2. Create plan to migrate us to more mobile friendly data vis platform; and

Phase 3. Migrate us. 

In your proposal please share examples of visualizations, details on specific Tibco Spotfire capabilities as well as Data Visualization capabilities.

Objective: We have panel data which other companies use to conduct surveys. We need to segment the panelists based on what motivates them to respond - incentive ($ paid for responding to surveys), survey length, topic, panelists demographics, psychographics etc. We need to surface meaningful segments in context of survey responses rate which be used to create segment level rate card. For example, segment 1 could be purely driven incentive while segment 2 is more about contributing to the society and does not care about incentive as along as the length of the survey is under x.

Data:

1. We have longitudinal data about the panelists including which survey they received, incentive, etc.
2. We have whether the panelists opened the survey invitation, started and or completed it
3. Panel member demographics, psychographics etc.

Technique: Looking for someone who specializes in surfacing segments using both categorical, numerical, and mixed type about variables on longitudinal data. Someone who specializes in latent class analysis using latent gold is highly preferred.

Timeline: Starts immediately. Segments to be delivered within a week.

When submitting the proposal, please let us know if you are expert in latent class segmentation or not and share specific details on projects executed using latent class segmentation.

Objective: We would like to extract features and segments from the histology images. The features include size of the cell, area etc. We will first have 100 images for training and source more (unto 500 if required) to build the feature extraction and segmentation deep learning algorithm.

Output: The algorithm (CNN or in combination with any other) will be able to extract features and segment areas via algorithm that would be consumed through batch processing interface. You will need to provide details of algorithms, their accuracies, etc. so that we can validate the algorithm.

In your proposal tell us more about your experience with feature extraction from images and deep learning expertise. Please provide methodology and fixed cost price in proposal, i.e. no of hours and price/hour breakdown along with milestones.

This project will be awarded to pre-selected expert.

Objective: Perform segmentation on integrated data, build typing tool, and build predictive scoring algorithm to score internal database among identified segments. 

Validation: The segments should have meaningful differences by survey data and most importantly internal transaction data. The predictive scoring algorithm should be highly accurate (74%+) 

Please do not bid on this project as this will be awared to pre-selected expert.

Goal:

Find market size data for Ground Vibration Monitors, IoT Tilt Meters, and IoT Crack Sensors.  It can be in applications in any sectors.  Particularly in Construction, skyscrapers, manufacturing, HVAC, Oil, Mining, Subway train systems, healthcare, and other sectors.

 

About us:

We provide a complete solution for remote construction site monitoring of vibration,noise and geotechnical parameters.

Our sensors provide an insight into the implications of vibrations for everyone. We developed measurement tools that are easy to use and that comply with all legal standards and quality requirements, to take accurate measurements in accordance with the applicable guidelines.

 

Deliverables:

• We need 2016 and 2017 as sold data in units and revenue.   We also need unit and revenue projections for 2018, 2019, and 2020.
• We need data on average selling price and number of units by the monitor type.
• We need 2016 an d 2017 of revenue and unit actuals of competitors Instantel, humboldtmfg, and sigicom
• We need services revenue actuals of 2016 and 2017.  We need services revenue projections for 2018, 2019, and 2020.
• We need the revenue split between North America, South America, Europe, Asia, and Australia/New Zealand
• We would be interested in perspectives on new use cases for vibration, tilt and crack monitors/sensors/meters
• We would be interested in actuals of 2016 and 2017 and future revenue projections for federal government/ state governments infrastructre building  projections 
• What is the  ideal customer profile?

 

Current Competition 

https://www.humboldtmfg.com http://www.instantel.com/  http://www.sigicom.com/ www.bksv.com  Lord Sensing, RST, Geokon, Slope Indicator, Sencieve*, sisgeo, wizen, encardio-Rite, and Logic technologies (Global sensing solutions). 

Enclosed case study and potential customers.

Details: Perform segmentation on consumer data (behavior, attitude, perception, and usage). 

Deliverable: Segmentation schema and typing tool. Detailed methodology. 

 Please do not bid on this project as this will be awared to pre-selected expert.

Our most pressing problem is the generation of a immune cell network, combining several -omics data sets to examine cellular interactions.

Project Goal:  The project will combine three different biological measurement types examining protein and gene expression on populations and individual cells, use this data to examine how transcriptional control can be examined as “internal sentiments” affecting the protein expression (translational control) and thus interaction between cells in different states (resting, activated, diseased).  Thus, the project will combine transcriptome, proteome, and interactome data -- creating a model of steady state of these in order to examine, and predict the effect of perturbations in the larger immune cell network.  

Research Project Summary: We will generate a framework for immune health by creating a social network of immune cells, modeling both internal state and interactions with other cells. To do this, we will analyze and combine data sets including (1) fluorescence-activated cell sorted cell subsets (FACS) analyzed with high-resolution mass-spectrometry-based proteomics and (2) fluorescence-activated cell sorted cell subsets (FACS) whose whole transcriptome has been analyzed. With (truly) quantitative protein and gene expression data for 28 human hematopoietic cell populations, and more than 10,000 proteins and genes measurements per cell, we generate an interactome for these cells and layer in both transcriptional and translational controls. Using these as internal ‘sentiments’, we will develop a deep steady state social network such that perturbations to the internal state of individual cells (e.g. in the context of disease) can be modeled. Using three examples, (1) CD8 T cell exhaustion, (2) alternate macrophage activation, and (3) dendritic cell dysfunction, we show how perturbations in immune cell gene and protein expression affect the network of immune cells, and ultimately the immune system.

Business Summary: We provide the leading analysis platforms for single cell analysis. 

Expertise Required: 

The ideal candidate or candidate organization will have experience with bioinformatics and a deep understanding of either genomics or proteomics data analysis.  Preference will be given to those candidates with a demonstrated ability to produce high-quality visualizations, experience with networks, graph-based approaches, and machine learning. Candidates with deep lateral experience in social network modeling and analysis but who lack the requisite domain experience will be considered.  

• Demonstrated ability to produce modern, interactive data visualizations (d3.js or equivalent)
• Demonstrated abillity to analyze large biological data sets (R and/or python)
• understanding and experience with bioinformatics analysis (data reduction, Pearson correlation, statistics)
• Demonstrated experience with machine learning (decision trees, ensembles, Logistic Regression)

Data Sources:

Data will primarily be from: Rieckmann et. al. Social network architecture of human immune cells unveiled by quantitative proteomics.  Nature Immunology volume 18, pages 583–593 (2017).  Also available at https://www.nature.com/articles/ni.3693 which includes supplemental information.  (the sample project file is Rieckmann et. al. Supplementary Table 3)

Zheng et. al. Nat Commun. 2017 Jan 16;8:14049 is the source of most of the single cell gene expression data sets.

 

Deliverables: 

There will be three phases of deliverables.  At each sub-step, the deliverable is generally a high-quality visualization as well as derivative data and source for producing the visualization (e.g. scripts).  No integration into our current technology stack is desired.  

Working Model:

• A detailed scope of work which is specified at the level of a requirements document will be provided to prospective candidate or candidate teams.
• Bi-weekly "standup" calls will be established to track progress on the project. 
• Project will be completed in three phases over four weeks by mid-April.  

Goal:

We are a Fortune 500 company in the Healthcare sector. We want to develop software or software and a device that will capture audio and save it to a non-audio format.  We want this software/device to capture data about audio that can be used for machine/deep learning projects.  The need to protect patient privacy is the key driver of this project’s main requirement – save audio data to a non-audio file.

Our Ideal Expert:

Someone with extensive experience programming audio software and/or someone with a strong understanding of acoustics – for example, an Acoustical Engineer or someone with expertise in voice analytics.

The Requirements:

Our initial thought is to develop software that would run on a Raspberry Pi.  This is not a hard requirement, but it would be a useful starting point for a prototype.  We are open to an expert’s advice if you want to suggest off-the-shelf hardware that could be used.

The software developed will capture audio using an omni directional microphone.  The captured audio will not be written to hard disk in an audio file format.  Instead the attributes of the audio listed below, will be written to a CSV file.  Please note, the audio features listed below are only the minimum required features.  We will default to the expert’s advice for additional features to capture from a voice analytics perspective.

Features:

• Mel Frequency cepstrum coefficients
• Spectral slope
• Audio spectral flatness
• Audio spectral centroid
• Audio spectral envelope
• Any additional audio features that may help in the identification of sound

The CSV file’s rows will represent time increments (e.g. one second) and should be written as a timestamp. The file’s columns will be the audio features listed above.  An important requirement is that we as users of the software/device be able to adjust the increments for sound capture.  We also need to be able to adjust the intervals in which the data is captured.  For example, we may want to set the device to record one second of sound every two minutes.

The software/device will need the ability to connect to a Wi-Fi network and transfer data collected to a destination specified by the user. The user will also need the ability to define the interval of data transfer.

Finally, the data recorded by this device should never be able to be converted into sound (i.e. audio could not be determined from the data captured).

At the end of the project the expert should produce full specifications for the hardware device and microphone used in development.

Proposals:

In your proposal please provide the following:

1. Your suggestions on hardware and your specific approach to this problem.
2. How will you ensure that the audio data is converted to delimited file and that no audio file is created in the process?
3. What are your proposed milestones and how many hours should be budgeted for each?
4. Your relevant experience with a project like this one.

TASK

Create the list of potential data sources and conduct feasibility assessment of these data sources to determine if and how data science can be applied (separately and together) to grow revenue and reduce costs for retailers and consumer goods manufacturers.

 

 

COMPANY OVERVIEW

INS builds a scalable blockchain-based Direct-to-Consumer mass personalization platform for the consumer sector (FMCG, CPG, groceries). There are 3 involved parties: manufacturers, retailers, and consumers. Consumer goods manufacturer and retailers will be able to provide personalised offers directly to consumers (discounts, promotions, etc.) via mobile app based on consumer personal data (gathered online), consumer purchase history (gathered offline via POS on checkout), and other data sources (from manufacturers, retailers, etc).

 

 

PROBLEMS TO SOLVE

•          What data should be gathered?

•          How to use these data to increase sales in retail stores?

•          How to use these data to increase sales for selected consumer goods manufacturers?

•          How to align interests of consumer goods manufacturers and retailers?

•          How to make personalised offers (discounts, rewards, etc.) for consumers with best-in-class personalisation and success rate (customers view offers via a mobile app and purchase promoted items in offline stores)

 

 

WHO WE NEED

•          Data scientist with 3+ years of experience in building mass market products

•          Deep understanding of the retail sector

•          Deep understanding of the consumer goods sector

•          Out-of-the-box and creative mindset

 

 

DATA SOURCES TO BE USED

•          Consumer personal data

•          Consumer social networks data

•          Data from a consumer goods manufacturer

•          Data from a retail store

•          Any other data sources that can be useful (weather forecast, holidays schedule, traffic situation, location-specific data, ect)

 

 

TECHNOLOGY

•          Ruby on Rails

•          Hadoop

•          Any other relevant technologies (we're open to the most advanced and cutting-edge solutions; we haven't started the development yet)

 

 

DELIVERABLE

Advisory service on developing a big data strategy, architecture and roadmap.

Background: We have build need-based segments using survey data of b2b customers sourced from thousands of internal b2b database. There are six segments.

GOAL: Build predictive model to classify segment membership of all the b2b customers in the database (who were not part of the survey).

Data:

1. Survey data with segments already identified.
2. We will provide 2 years of internal transaction data. See sample of variables captured in the attached file. Output: Classification model to classify segment membership of ~29k B2B customers into one of the six segments. They need the code to run internally. 

Your task:

1. Understand secondary data
2. Build predictive model to classify internal customers to one of six segments with probabilities of being into each segment
3. Provide code to client so that they can run it internally. The type of code will depend on client’s system (most likely python or sql etc. )

Please do not bid on this project as this will be awared to pre-selected expert.