Who We Are
Girl Effect is an international non-profit that builds media that girls want, trust and need.
From chatbots to chat shows and TV dramas to tech, our content helps adolescent girls in Africa and Asia make choices and changes in their lives, igniting their confidence to act differently at a time that can define their future.
We create safe spaces for girls, sharing facts, and answering questions about health, nutrition, education, relationships, and so much more. To arm girls with the skills to negotiate and redefine what they are told is possible “for a girl”.
Our reach is 50 million and counting. And we’re using technology to reach girls at scale so every girl can choose to be in control of her body, health, learning and livelihood.
Because when a girl unlocks her power to make different choices that change her life, it inspires others to do so too. She starts a ripple effect that impacts her family, community, and country.
That’s the Girl Effect.
Background
For over a decade, we have been building technology that aims to create a safer, more accessible world for women and girls, using the insights and experiences girls share with us. Innovation for us is about using the right tech in the best way to reach girls and meet their challenges. As technology evolves and girls’ habits change, we adapt and improve the platforms, processes, and infrastructure that underpin everything we do.
One of our flagship products is a chatbot tailored for young girls in developing countries, embodying the persona of a caring big sister. Currently active in South Africa, India, and Kenya, these chatbots converse in the mixed-coded, youth-centric languages unique to each region, mirroring the linguistic nuances of the girls they assist.
Context
Girl Effect’s chatbots today are menu-based with classification systems integrated for natural language understanding, such as Big Sis, our WhatsApp chatbot in South Africa. These classification systems enable Girl Effect’s chatbots to handle freeform input and direct the user to the most relevant message set based on the category detected but do not allow Girl Effect’s chatbots to generate answers that directly address user’s freeform input.
Recent developments in GenAI highlight the limitations of classification intelligence and, at the same time, expose the potential to unlock opportunities that were once considered impossible. We believe we can overcome the limitations of our classification based chatbots and pave the way for the coming wave of transformative technology, ensure that it will be designed to address the needs of the most marginalised and make sure that both we and the girls we aim to support will not be left behind.
To fully harness the immense potential of GenAI and avoid being left behind, Girl Effect has an ambitious roadmap. Part of this roadmap includes research and development towards a proof of concept of Gen AI enhanced conversational chatbot focusing on knowledge transfer and intent to take action. The work will explore the following hypothesis: using GenAI, we can develop conversations that understand & speak low-resource languages and that accurately and safely inform young people of their sexual reproductive health and rights, mental well-being and how to access health services when ready.
Further personalization and scaling the dialog flow for a public release are not included in the scope of this workstream.
We have allocated 6 months to complete this workstream with the requirements set forth below.
Objectives
The aim is to develop a proof-of-concept chatbot that can handle open-ended user inquiries using Girl Effect content developed by a health and gender team. This content will draw from years of Q&A sessions with real girls in our target markets. This content, combined with data on local services such as health clinics or mental health services, will help equip young people, when ready, to access these resources, ensuring their physical and emotional needs are met.
Within this piece of work, we’ll set up the foundational infrastructure to support and possibly deploy the chatbot. It’s important to note that scaling this infrastructure and releasing the chatbot to the public aren’t included in this phase. Instead, we’ll rely on internal users and focus groups for evaluation and testing to validate our proof of concept.
Key Deadlines
We will be taking a phased approach to delivering conversational AI. This TOR defines the first Phase to be delivered within 6 months of the project start.
Within this 6 month deliverable, we will focus on a proof of concept to ensure that we can provide a relevant, safe and engaging conversation chatbot. For this phase, we have identified three key learning questions that will inform our understanding of the viability of available vendor and open source models to perform the specific capabilities to the performance standards required by Girl Effect.
- Tone, language and Relevance: Can the latest technologies in NLP and Large Language Models deliver a highly personalised experience in tonality, language and relevance and enhance personalised experience by directly providing users with the content they request in a tone and language that meets their cultural preferences?
- Safeguarding: Can the latest technologies continue to provide a safe environment for our users and enhance our ability to detect sensitive disclosures and direct users to resources and services when they need them?
- User Engagement: Will integrating Gen AI and switching from a menu-based experience to a conversationally-led experience enhance our user engagement?
Deliverables
You will deliver a proof-of-concept chatbot that can handle open-ended user inquiries using Girl Effect-approved content, vetted GenAI models, and chatbot and infrastructure frameworks.
We aim to rigorously evaluate various open-source and proprietary language models. Our research methodology will be meticulously crafted to address each learning objective, fulfill all requirements, and encompass the entire project scope.
We will handpick models for testing based on their proven efficacy, general recognition in the field, and substantiated outcomes. An initial set of models will be established at the project’s outset, anticipating iterative refinements over the six months, influenced by emerging insights and model availability.
The vendor will employ an evidence-based testing framework to assess each model’s Accuracy, Efficiency, and Cohesiveness, considering factors like tone, language adaptability, and contextual relevance. We expect the vendor to implement a suite of automated tests, enabling precise and consistent comparative analysis of each model’s performance.
Your role also entails delivering expert recommendations for model hosting, maintenance infrastructure, chatbot frameworks, and the necessary supporting infrastructure. You are to utilize evidence-based methods for all evaluations. You will be accountable for identifying and detailing all associated costs for each recommended framework and infrastructure solution, whether Cloud, SaaS, or other related expenses.
Requirements
- Chatbot Framework: It’s essential to evaluate the chatbot platform and framework to choose the best-fit solution. You will be responsible for accessing and selecting the appropriate framework for our next-generation conversational GenAI chatbots.
- Vetted Content: Ensuring our chatbots disseminate exclusively verified information is crucial. Your task is to build a chatbot infrastructure that guarantees user responses are confined to content sanctioned by Girl Effect, encompassing approved knowledge and service-related data. Employ and rigorously test suitable methodologies to ensure the model delivers safe and vetted content consistently.
- Technical safeguarding: The infrastructure must incorporate mechanisms for identifying safeguarding disclosures and initiating the correct response protocols. The proof of concept (POC) needs to showcase the ability to recognize safeguarding instances and seamlessly integrate them into the appropriate action-oriented workflows.
- Mix-code Languages: Girl Effect operates in areas where low-resource languages prevail. (languages that lack large monolingual or manually crafted linguistic resources sufficient for building statistical NLP applications). Not only are users in these regions speaking in low-resource languages, they also often mix languages and local slang, speaking in what is termed mix-coded languages. Relevant techniques should be employed and evaluated to facilitate conversations in specific mixed-code languages. Specifically in scope will be Hinglish for the Indian market and Sheng for Kenya.
- Culture and Tone: The design of Girl Effect’s chatbot user experience must mirror the local cultural context and resonate with the user’s tone. Employ and evaluate effective techniques to capture the youthful essence that aligns with the Girl Effect brand identity.
- Speech to Text and Text to Speech: Understanding the capability of speech-to-text and text-to-speech chatbots for integration into IVR or other lo-fi systems for processing and communication through voice notes. This scope should include testing speech-to-text and text-to-speech for the specified low-resource and mix-coded languages.
- Production infrastructure: Based on the POCs implemented in this workstream, a production infrastructure should be created capable of supporting and testing the relevant models and chatbot infrastructure.
At the end of the 6 months, we want a working prototype. However, that isn’t the only end goal of the workstream. We also want to develop our understanding of how we can integrate into these systems to plan for a longer-term chatbot and generative AI infrastructure grounded in the most cutting-edge technologies.
Expected commitment:
- The agency will pursue this workstream over 6 months.
- Girl Effect and the consulting agency will work as one team, fully integrated with regular check-ins.
Resource Requirements
Girl Effect will provide supervision and specific areas of expertise. While we will entertain supplemental resources in those areas, we expect the consulting agency to provide the remaining resources.
Girl Effect provided resources:
- Project Management – Girl Effect will provide a dedicated project manager
- Content Specialist – Girl Effect will provide the content for the targeted markets and the content specialists within those markets
Consulting agency provided resources:
In your proposal, please make sure to include at least the specified resources outlined below.
- ML Engineers – The ML engineers will have experience in generative AI and large language models. The consulting agency will provide the ML engineers with the expertise level needed to accomplish the workstream scope.
- MLOps – The consulting agency will be responsible for all MLOps and infrastructure setup.
- Chatbot Engineer – We expect the need for at least one chatbot engineer to explore different chatbot frameworks and develop the front-end interface for the chatbot.
- Quality Assurance – The vendor will be responsible for quality assurance. We expect the vendor to implement automation strategies to streamline the testing process and ensure consistent results across model iterations.
Our Expectations
In our terms of service, we’ve detailed specific techniques and requirements. We anticipate that your team and the resources you provide will meet these expectations. Beyond just meeting requirements, we value proactive thought leadership. We desire your consulting agency and its team members to be at the forefront of the generative AI field, always updated and open to suggesting innovative approaches and methodologies. To us, success is not solely a functional POC. It encompasses a partnership and collaboration where together we pioneer new frontiers, generating rich learnings and insights which, in turn, enable the creation of products that have a meaningful impact on the lives of young people.
Location
The consulting agency will be remote and must be conducive to working with teams in the UK, USA, Sub-Saharan Africa & India synchronously and asynchronously.
Tentative Timelines
- Terms of reference published: November 22, 2023.
- Deadline for Questions/Clarifications: November 28, 2023
- Deadline for responses from GE:December 1, 2023
- Proposal submission: December 8, 2023
- Project commencement: December 18, 2023
Reporting
The consulting agency will report to Girl Effect’s Chief Creative and Technology Officer. They will work closely with the Fractional CTO and Digital Data Analyst.
Who You Are
You are an agency with a wide range of skills and expertise in:
- Natural Language Processing (NLP) development
- Coding for different language detection (incl. low-resource languages) and translation
- AI model building and optimisation
- Experience in fine-tuning language models for tone
- Prompt engineering experience
- Chatbot engineering and architecture
- Familiarity with a range of NLP software and experience with integration testing
- Agile methodologies and the software development lifecycle
- Familiarity with BERT and other large language models, including open source models
- Familiarity with Hugging Face
- Programming languages such as Python
- Machine learning frameworks such as TensorFlow, PyTorch, and scikit-learn.
- Deep learning architectures such as RNNs, LSTMs, CNNs, and transformers
- Pandas, NumPy, and SQL for data set preprocessing and analysis
- Linguistic theories and concepts, such as syntax, semantics, and pragmatics
- Various machine learning techniques (Regression, Clustering, Decision Tree, Probability Networks, Neural Networks, Bayesian models etc.).
- Experience working in Low- or Middle-Income Country contexts
Proposal submission
Your proposal should be 5 pages max and cover the following:
- Your understanding of the brief and why you feel you are well-placed to advise us on this
- Credentials/Profile/CV – Please showcase any previous work that you consider relevant to what we are trying to achieve, how we approach our work, and the audience we cater to. Describe the qualifications, experience, and capabilities of the firm or consultant in providing the requested services.
- Resource Model: Please provide detailed specifications on the number, variety, and skill sets of each team member you will allocate to this project.
- Reference- Provide at least three references for similar contracts with a description of the service provided, the value of the contract, and the contract periods of performance.
- Breakdown of Costs with applicable taxes and other charges clearly identified in USD. Consultants should provide their daily or hourly billing rate or their proposed fixed fee for the requested deliverables.
- All applicable taxes should be quoted separately;
In their proposal, the bidder must demonstrate an understanding of the requirements described in the RFP and demonstrate how the bidder will meet the requirements of the evaluation criteria.
GE is not liable for any cost incurred during the award/contract preparation, submission, or negotiation of the award/contract. All submitted documentation and/or materials shall become and remain the property of GE.
VALIDITY of the proposal shall be for 90 days from the date of bid closure
Evaluation Criteria
The criteria against which proposals will be evaluated are listed below.
Technical Evaluation
- Interpretation and understanding of this TOR – 10%
- Showcases relevant experience in having done previous similar work with relevant agencies or organizations – 10 %
- Demonstrates familiarity with the artificial intelligence sector in the geographies that we operate in – 10%
- Familiarity with technologies and market trends that will impact artificial intelligence/Machine learning in the future and organizations that are to be expected in this space-15%
- Evidence of a minimum of three contactable references – 10%
- Experience and Qualification, share your CV/Profile – 15%
Financial Evaluation
- Value for money/proposed budget breakdown – 30%
Questions/Clarifications
If you have any questions about this RFP, please email suppliers@girleffect.org by November 28, 2023. All questions will be answered and shared through an FAQ on our website.
Equal Opportunities
Girl Effect is committed to equal opportunity regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, marital status, disability, gender, gender identity, or expression. We are proud to be an equal-opportunity workplace.
We are committed to building an organization that is truly representative of the communities that we serve. To this end, due regard will be paid to procuring consultancy services of organizations and individuals with a deep understanding and experience of our programming markets and diverse professional, academic and cultural backgrounds.
Safeguarding
You may be required to undertake safeguarding checks. Shortlisted consultants will be assessed on our organizational values at the interview stage. The successful consultant will be expected to adhere to our safeguarding policy. We encourage you to read and understand our safeguarding policy, the executive summary of which can be found here. We have zero tolerance for all forms of violence against children, beneficiaries and staff.
Tax
Applicants are advised to ensure that they have a clear understanding of their tax position with regards to provisions of their local jurisdiction tax legislation when developing their proposals.
Disclaimer
GE reserves the right to determine the structure of the process, number of short-listed participants, the right to withdraw from the proposal process, the right to change this timetable at any time without notice and reserves the right to withdraw this tender at any time, without prior notice and without liability to compensate and/or reimburse any party. GE shall inform the ONLY successful applicant(s). The process of negotiation and signing of the contract with the successful applicant(s) will follow.
How to apply
Please submit proposals, as described above, to suppliers@girleffect.org by December 8, 2023, 5:00 pm EAT latest. Please clearly mark your email with the subject ‘Generative AI Conversational Chatbot’.