Submitted
Last Updated August 10, 2018
Work of the Future
iMerit Technology Services
Team Leader
Jai Natarajan
Basic Information
Our tagline:
Hiring & training global low income communities in data & content work for Machine Learning and Computer Vision
Our pitch:
The profit pool in machine learning and computer vision is directly driven by the availability of large differentiated data sets. The requirement for data increases in far greater proportion (~2x previous set) compared to marginal increase in performance, over iterations. On-target quality data with relevant input parameters works better than non-targeted data to enhance algorithm performance. Creating this data through conventional means is prohibitively expensive. Meanwhile there is a large pool of educated but digitally excluded workforce seeking opportunities.
iMerit follows a unique for-profit social impact model where we hire and train underprivileged employees in these data enrichment and labeling skills assisted by technology tools and domain skilling. They allow us to provide an enterprise solution at scale and quality, for the data-hungry customers.
The partnership benefits from social inclusion, diversity, scalability and motivation of the workforce. The combination of human judgement, tools and training has been and continues to be a value creating approach for our clients. Our cloud based tools and seamless skilling approach, allow the work to be done from historically neglected geographies like Kolkata and New Orleans. 80% of employees come from poor families and 50% are women. All are transforming their communities and family fortunes. Hence we achieve a perfect storm of digital inclusion, social impact, diversity and aspirational livelihoods in a sector often accused of destroying jobs. We are building the foundation for human expertise to collaborate with machine power to provide nuanced AI solutions at scale.
iMerit follows a unique for-profit social impact model where we hire and train underprivileged employees in these data enrichment and labeling skills assisted by technology tools and domain skilling. They allow us to provide an enterprise solution at scale and quality, for the data-hungry customers.
The partnership benefits from social inclusion, diversity, scalability and motivation of the workforce. The combination of human judgement, tools and training has been and continues to be a value creating approach for our clients. Our cloud based tools and seamless skilling approach, allow the work to be done from historically neglected geographies like Kolkata and New Orleans. 80% of employees come from poor families and 50% are women. All are transforming their communities and family fortunes. Hence we achieve a perfect storm of digital inclusion, social impact, diversity and aspirational livelihoods in a sector often accused of destroying jobs. We are building the foundation for human expertise to collaborate with machine power to provide nuanced AI solutions at scale.
Where our solution team is headquartered or located:
Kolkata, India
The dimensions of the Challenge our solution addresses:
- Other (Please Explain Below)
- Upskilling, Reskilling, and Job Matching
About Your Solution
What makes our solution innovative:
1. FigureEight - pioneer in crowdsourced data labeling tasks; focused on individual anonymous crowd versus enterprise partnering. Strong AI capability
2.MightyAI - a task platform for labeling work. Tehcnology focused rather than workforce and skill focused
3. CloudFactory - similar to iMerit but with more of a backoffice focus than Machine Learning. Powerful impact model with mix of inhouse and mostly contract remote workers
Imerit competitive advantage - impact sourcing for quality and cost advantage; technology tools for skilling and delivery; fulltime employment model; enterprise white glove service
2.MightyAI - a task platform for labeling work. Tehcnology focused rather than workforce and skill focused
3. CloudFactory - similar to iMerit but with more of a backoffice focus than Machine Learning. Powerful impact model with mix of inhouse and mostly contract remote workers
Imerit competitive advantage - impact sourcing for quality and cost advantage; technology tools for skilling and delivery; fulltime employment model; enterprise white glove service
How technology is integral to our solution:
1) Cloud-based service delivery platform which manages all resources and projects seamlessly across all the delivery centers. Throughput and quality are managed through dashboards. Can connect to third party platforms via APIs.
2) Individual task user interface for the worker to interactively perform the annotations. Tasks may involve drawing bounding boxes around all vehicles and pedestrians within images of street scenes
3) Simulation versions of the tools are used for skilling workers, either as incoming trainees or as upskilling workers for new projects
2) Individual task user interface for the worker to interactively perform the annotations. Tasks may involve drawing bounding boxes around all vehicles and pedestrians within images of street scenes
3) Simulation versions of the tools are used for skilling workers, either as incoming trainees or as upskilling workers for new projects
Our solution goals over the next 12 months:
Metrics:
1. Number of impact employees
2. % of impact employees (family income < $200 per month)
3. % of women
4. Income improvement in impact employees
5. Revenues
6. Average daily revenue per employee
7. Number of team leads from impact backgrounds
Goals:
1. Achieve USD 50 Million in global revenues / 10,000 employees
2. Maintain over 60% impact employees
3. Achieve profitability at net profit level
1. Number of impact employees
2. % of impact employees (family income < $200 per month)
3. % of women
4. Income improvement in impact employees
5. Revenues
6. Average daily revenue per employee
7. Number of team leads from impact backgrounds
Goals:
1. Achieve USD 50 Million in global revenues / 10,000 employees
2. Maintain over 60% impact employees
3. Achieve profitability at net profit level
Our vision over the next three to five years to grow and scale our solution to affect the lives of more people:
Our vision for the organisation is an insightful and expert data partner for machine learning companies which provides a unique point of view and value proposition due to its model of digital inclusion and social impact.
We stand directly in contradiction of the current streams of anxiety which relate to job destruction, digital inequality, opportunity stagnation and non-diversity, in the cutting edge of the technology sector.
We stand directly in contradiction of the current streams of anxiety which relate to job destruction, digital inequality, opportunity stagnation and non-diversity, in the cutting edge of the technology sector.
Our website
https://imerit.net
Our promotional video:
Find out more about us. First link:
https://www.youtube.com/watch?v=_Fmcot-jWuk&t=4s
The regions where we will be operating in the next 12 months:
- South Asia
How we will reach and retain our customers or beneficiaries:
In our New Orleans center several immigrants and climate refugees from Puerto Rico are using their multi-lingual skills in a digital environment.
We provide a technology enabled enterprise environment to partner with customers and give them the comfort that their projects are in secure hands with highly motivated and skilled employees.
We provide a technology enabled enterprise environment to partner with customers and give them the comfort that their projects are in secure hands with highly motivated and skilled employees.
How many people we are currently serving with our solution:
Per last board meeting: 1300 jobs created; Impact Ratio 80%; Diversity - 48% women; Team leads promoted from impact background: 30+;
External Report - http://gcyerti.com/wp-content/uploads/2017/02/The-Nexus-of-Microwork-and-Impact-Sourcing_Final_ONLINE_02.28.17_v2.pdf
Silver Stevie Award 2017 for Fostering Inclusion and Diversity - https://www.youtube.com/watch?v=WtD34SBjtu4
https://imerit.net/imerit-ceo-honored-omidyar-networks-entrepreneurs-changing-education/
Press:
https://www.hindustantimes.com/india-news/big-data-to-cameras-how-technology-is-empowering-muslim-women/story-Rmu0IuX9zpIiQaxekYU2hN.html
https://www.washingtonpost.com/national/ai-has-a-dirty-little-secret-its-powered-by-people/2018/03/05/53e20b70-208f-11e8-946c-9420060cb7bd_story.html?utm_term=.3adff994bedf&noredirect=on
https://www.msdf.org/blog/2016/04/invested-in-imerit/
http://observer.com/2016/11/how-the-chicago-cubs-used-big-data-to-win-the-world-series/
External Report - http://gcyerti.com/wp-content/uploads/2017/02/The-Nexus-of-Microwork-and-Impact-Sourcing_Final_ONLINE_02.28.17_v2.pdf
Silver Stevie Award 2017 for Fostering Inclusion and Diversity - https://www.youtube.com/watch?v=WtD34SBjtu4
https://imerit.net/imerit-ceo-honored-omidyar-networks-entrepreneurs-changing-education/
Press:
https://www.hindustantimes.com/india-news/big-data-to-cameras-how-technology-is-empowering-muslim-women/story-Rmu0IuX9zpIiQaxekYU2hN.html
https://www.washingtonpost.com/national/ai-has-a-dirty-little-secret-its-powered-by-people/2018/03/05/53e20b70-208f-11e8-946c-9420060cb7bd_story.html?utm_term=.3adff994bedf&noredirect=on
https://www.msdf.org/blog/2016/04/invested-in-imerit/
http://observer.com/2016/11/how-the-chicago-cubs-used-big-data-to-win-the-world-series/
About Your Team
How our solution team is organized:
- For-Profit
Explaining our organization:
iMerit was set up in 2012 with its first operations in Kolkata in Metiabruz, a highly conservative Muslim neighborhood where iMerit set up an all-women's center. A US subsidiary followed with sales and executive leadership based there. As of March 2018 iMerit has 6 delivery centers (three in Kolkata, plus Ranchi, Bhubaneshwar and New Orleans), with 1300+ employees. 80% of employees come from family incomes of under USD 100 per month. 50% are women. Over 90% of customers are in the US including Fortune 5 retail, Financial information services giant, and several hot startups in Machine learning and Computer Vision.
How many people work on our solution team:
- 20+
How many years we have been working on our solution:
- 5-10 years
The skills our solution team has that will enable us to attract the different resources needed to succeed and make an impact:
Radha Basu, founder & CEO--pioneer in the software business who broke the glass ceiling in Silicon Valley.
Built HP's software division to a $1.2 billion business and launched HP in India.
Former CEO of Support.com, which she led to IPO on Nasdaq. Launched Anudip Foundation in 2007 to generate livelihoods for India's marginalized youth through digital and workplace skills development.
Awards:
-Silicon Valley Business Journal Women of Influence Award
-the UN Women-ITU Gender-Equality Mainstreaming Technology Award
-Top 25 Women of the Web, CEO of the Year and in 2018
-the GTF (Global Thinkers Forum) Award for Excellence in Youth Development
S
Built HP's software division to a $1.2 billion business and launched HP in India.
Former CEO of Support.com, which she led to IPO on Nasdaq. Launched Anudip Foundation in 2007 to generate livelihoods for India's marginalized youth through digital and workplace skills development.
Awards:
-Silicon Valley Business Journal Women of Influence Award
-the UN Women-ITU Gender-Equality Mainstreaming Technology Award
-Top 25 Women of the Web, CEO of the Year and in 2018
-the GTF (Global Thinkers Forum) Award for Excellence in Youth Development
S
Our revenue model:
iMerit has both social objectives and business objectives in our Articles of Association, creating a unique social impact for-profit model. Our strategy is to seek areas of global services where we can train and deploy educated employees from poor backgrounds.
iMerit has scaled to 1300+ employees and is expected to double this year. The employee capability ladder includes skills like image labeling, sentiment analysis, text labeling and content enrichment. Dynamic micro-skilling allows employees to quickly work on a new domain or use case. Proprietary software tools allow the efficient allocation and delivery of the data tasks. A US based sales and solutions team ensures a high standard of technological capability and close customer connect. Up to 40% of our staff can be from non-impact backgrounds (currently 20%). This includes executives, sales, marketing, and technology functions. Delivery staff can also grow into team leads and delivery managers. Our daily average revenue per employee has grown each year (doubled in three years) due to the increase in skill and productivity levels, better use of our enterprise service delivery platform, and continuous upskilling of the team. We are evolving into trusted partners providing insights and best-practises knowhow to customers.
iMerit has scaled to 1300+ employees and is expected to double this year. The employee capability ladder includes skills like image labeling, sentiment analysis, text labeling and content enrichment. Dynamic micro-skilling allows employees to quickly work on a new domain or use case. Proprietary software tools allow the efficient allocation and delivery of the data tasks. A US based sales and solutions team ensures a high standard of technological capability and close customer connect. Up to 40% of our staff can be from non-impact backgrounds (currently 20%). This includes executives, sales, marketing, and technology functions. Delivery staff can also grow into team leads and delivery managers. Our daily average revenue per employee has grown each year (doubled in three years) due to the increase in skill and productivity levels, better use of our enterprise service delivery platform, and continuous upskilling of the team. We are evolving into trusted partners providing insights and best-practises knowhow to customers.
Partnership Potential
Why we are applying to Solve:
Financial support:
We want to use ML in our operations and skilling. This includes developing ML 'helpers' for the labeling work done by our workers which can enhance throughput and accuracy. We will use the Solve's support to grow an ML team in India specifically 2-3 ML engineers, 2-3 full stack developers and a product manager to integrate the resulting solutions in our platform. ML will also be used to assess the work of trainees done in training projects and predict the best for for their skillsets. Marketing will showcase this to clients as a differentiator.
We want to use ML in our operations and skilling. This includes developing ML 'helpers' for the labeling work done by our workers which can enhance throughput and accuracy. We will use the Solve's support to grow an ML team in India specifically 2-3 ML engineers, 2-3 full stack developers and a product manager to integrate the resulting solutions in our platform. ML will also be used to assess the work of trainees done in training projects and predict the best for for their skillsets. Marketing will showcase this to clients as a differentiator.
The key barriers for our solution:
Internal:
1. Saturation of capability ladder - in-house skilling team and diversification of talent pools from different geographies; domain skilling
2. business continuity - careful selection of sites with high speed internet, multiple redundant capacity, risk management (political/safety risks), spread of projects and skills across sites
3. Data security - following best practices, annual audits, ISO 27001 readiness
External:
1. Political - creation of US delivery center and US entity
2. Data regulations - creation of multiple locations; projects rarely depend on Personal Information
1. Saturation of capability ladder - in-house skilling team and diversification of talent pools from different geographies; domain skilling
2. business continuity - careful selection of sites with high speed internet, multiple redundant capacity, risk management (political/safety risks), spread of projects and skills across sites
3. Data security - following best practices, annual audits, ISO 27001 readiness
External:
1. Political - creation of US delivery center and US entity
2. Data regulations - creation of multiple locations; projects rarely depend on Personal Information
The types of connections and partnerships we would be most interested in if we became Solvers:
- Other (Please Explain Below)
Solution Team:
Jai Natarajan