Meta Global AI Developer Hackathon. $1M in prizes.Get notified
LLAMA 3.1 70B INSTRUCT
Instituto PROA
6x faster
Job dossier creation time dropped from 30 minutes to 5.Zero staff time
Students create job dossiers with self-service AI.60x program growth
Automation helped increase enrollment to 35K students a year.*All results are self-reported and not identifiably repeatable. Generally, expected individual results will differ.
CASE STUDY
Helping thousands of students automate job research and start their careers
At a glance
Industry
EducationUse case
Automating job-opening research and reportingGoal
Free up PROA staff and enable exponential growth in students servedLlama version
3.1 70B InstructDeployment
Oracle Cloud Infrastructure (OCI), OCI Generative AI APITHEIR STORY
Helping low-income students hone their professional skills
Instituto PROA helps young people from public schools develop personal and professional skills so they can become protagonists of their own stories as they enter the job market. Since 2007, PROA has reached more than 36,000 young people. Today, it operates in six of Brazil’s 26 states through its PROPROFESSION courses and the PROA Platform.THEIR GOAL
Overcoming job research bottlenecks with generative AI
PROA team members were creating dossiers to help high school students prepare for job interviews. But as PROA began to expand, hand-crafting each dossier based on the individual job opportunity became unsustainable.The team saw an opportunity to automate the production of research and job-candidate dossiers using generative AI. With the help of Oracle and Accenture, the PROA team set out to transform the process into a smart, efficient solution that could serve thousands of requests simultaneously.
THEIR SOLUTION
A Llama-powered self-service job research and dossier assistant
Using OCI Generative AI, PROA combined Llama 3.1 70B Instruct with a retrieval augmented generation (RAG) pipeline to create a self-service job research and dossier assistant.PROA chose Llama for its seamless integration with their existing OCI instance and strong performance for the dossier use case. Beyond that, managing costs was essential for the non-profit organization. Open-source Llama didn’t levy per-token inference fees, and it also brought the flexibility and innovation of the open-source community.
Students access the tool on PROA’s online platform and generate thoroughly researched dossiers in minutes.
THEIR APPROACH
OCI streamlined development and production
Development and production environments run in PROA’s OCI instance. When students request a dossier about a specific job or company, the PROA AI assistant calls a search engine result page (SERP) API, which reviews the public web for detailed information.The application combines the SERP API results with relevant context from PROA’s knowledge base and injects them into a RAG pipeline running on the OCI Generative AI. The Llama 3.1 70B Instruct model receives the context, original query and pre-engineered instructions, then generates a completed dossier for the student.
Privatemode AI verifies the integrity of each step of the confidential computing service.
THEIR SUCCESS
Llama-powered automation lays a foundation for expanded youth services
Automating dossier creation has sped up the process significantly and freed PROA team members to spend more time working directly with students. Through the online platform, the non-profit is using the assistant to extend professional development and career counseling beyond urban centers to young people across Brazil.
- 6x faster — Job dossier creation time dropped from 30 minutes to 5.
- Zero staff time — Students create job dossiers with self-service AI.
- 60x program growth — Automation helped increase enrollment to 35K students a year.
“We explored other LLMs, but Llama 3.1 was always our first choice due to its seamless integration with our Oracle Cloud Infrastructure and its strong performance for our dossier generation use case.”
Alini Dal Magro — Executive Director, Instituto PROA
“With the automated workflow, we can generate dossiers for multiple candidates simultaneously, significantly increasing our throughput. This scalability enables PROA to support more young people without compromising the quality or speed of our services, which aligns with our mission to maximize our impact.”
Alini Dal Magro — Executive Director, Instituto PROA
“By leveraging Llama 3.1 with RAG on OCI, we’ve transformed the dossier generation process into an automated, efficient and scalable workflow. What once took 30 minutes of manual effort now takes just 5 minutes, an 83% reduction in processing time, time our team can use for higher value tasks like candidate coaching.”
Alini Dal Magro — Executive Director, Instituto PROA
“PROA’s technological innovation has transformed the student experience and the organization itself. Now that it is known as a tech innovator, PROA is attracting more donations, partnerships and applicants.”
Alini Dal Magro — Executive Director, Instituto PROA
Previous slide
Next slide
Start building
Explore more
ConsumerBusiness
Shopify | Llama case studiesShopify uses Llama to generate product pages, localize content, and automate support, helping developers scale workflows and save time.Read more
ConsumerTech
Scribd, INC | Llama case studiesDelivering faster, cheaper and more accurate results with a Llama-powered AI content discovery assistant.Read more
Stay up-to-date
Our latest updates delivered to your inbox
Subscribe to our newsletter to keep up with the latest AI updates, releases and more.