Approximately a year ago, I led a significant national-level data curation initiative tasked with gathering local language data from 80 diverse districts across India. Our challenge was twofold: establish an efficient data collection and validation pipeline and swiftly assemble a capable team to execute this ambitious mission. Here’s how we accomplished it:
1. Securing Trustworthy Team Members:
To kickstart the program, I brought on board two reliable colleagues whom I had worked with closely in the past.
2. Strategic Promptness:
While we had vendors collecting data from various districts, we needed an in-house team of native speakers for data validation. Given the urgency and the extensive geography, we started engaging native speakers through freelancing and job search platforms. This allowed us to initiate the validation process without waiting to hire across all 80 districts.
3. Hiring Effective Freelancers:
We formed a core program team of 11, primarily composed of freelancers who exhibited exceptional performance and leadership skills during their engagement with the program.
4. Leveraging the Network Effect:
To scale the program further, we aimed to recruit at least 20 native speakers per district for regular validation work. Although AI models handled most validations, a small percentage required manual checks. We now assigned specific roles to each of the core team members: 3 managers & 8 associates. Each associate led 10 districts that they were most comfortable with in terms of local language familiarity. We designed a training program for them to recruit freelancers from their assigned districts. Our strategy harnessed the network effect, tapping into schools, colleges, and onboarding students and professors. If one professor expressed interest, we encouraged them to bring in other students and professors, creating a self-perpetuating recruitment cycle.
This approach enabled us to build a strong team of 1,000 within a remarkable two-month timeframe, ensuring the success of our program.