We present CAISSON, a novel hierarchical approach to Retrieval-Augmented Generation (RAG) that transforms traditional single-vector search into a multi-view clustering framework. At its core, CAISSON leverages dual Self-Organizing Maps (SOMs) to create complementary organizational views of the document space, where each view captures different aspects of document relationships through specialized embeddings. The first view processes combined text and metadata embeddings, while the second operates on metadata enriched with concept embeddings, enabling a comprehensive multi-view analysis that captures both fine-grained semantic relationships and high-level conceptual patterns. This dual-view approach enables more nuanced document discovery by combining evidence from different organizational perspectives. To evaluate CAISSON, we develop SynFAQA, a framework for generating synthetic financial analyst notes and question-answer pairs that systematically tests different aspects of information retrieval capabilities. Drawing on HotPotQA's methodology for constructing multi-step reasoning questions, SynFAQA generates controlled test cases where each question is paired with the set of notes containing its ground-truth answer, progressing from simple single-entity queries to complex multi-hop retrieval tasks involving multiple entities and concepts. Our experimental results demonstrate substantial improvements over both basic and enhanced RAG implementations, particularly for complex multi-entity queries, while maintaining practical response times suitable for interactive applications.

Debris from a drone strike in northern Ethiopia's Tigray region has damaged a truck carrying humanitarian aid and belonging to the World Food Programme (WFP) and injured the truck's driver, the United Nations agency said on Monday. The WFP said the drone strike on Sunday hit near an area called Zana Woreda in northwestern Tigray, as two trucks were delivering relief supplies to families displaced by the nearly two-year long conflict. "Flying debris from the strike injured a driver contracted by WFP and caused minor damage to a WFP fleet truck," the spokesperson said, adding it was not possible to say yet whether further distributions would be suspended in the area. "WFP calls on all parties to respect and adhere to international humanitarian laws and to commit to safeguarding humanitarian workers, premises and assets." The WFP truck was delivering food to internally displaced people as hundreds of thousands have been uprooted by renewed fighting since August 24 after a five-month ceasefire broke down.

Description:   *** Cannot provide sponsorship upon conversion. What is the specific title of the position? Lead Data and AI Engineer Work location? Preferred Locations - MA or MN (Client facilities). 100% telecommute is also considered. Work hours (ex. 9am-5pm day/night shifts rotating shifts etc)? 9-5 Please provide a summary of the project/initiative that this candidate will be working on? We are establishing Agile Data Warehouse in cloud and many new AI practices to enable personalization in various capabilities to improve employee experience. Please describe the team the candidate will be working with - how many members? 10 – 12 team members What is the break-down of the teams skill sets (ex: 1 PM 4 Developers etc.)? 1 PM, 3 Product owners 2 Sprint teams consisting - 1 Scrum Masters and 12 developers What are the top 5-10 responsibilities for this position (please be detailed as to what the candidate is expected to do or complete on a daily basis)? • Identify opportunities for Data Engineering and AI to enhance the core product platform, select the best machine learning techniques to the specific business problem and then build the models that solve the problem. • Architect and design AI/ML and Analytics solutions and cloud services • Own the end-end process, from recognizing the problem to implementing the solution. • Establish DataOps and MLOps principles and best practices What does the ideal candidate background look like (ex: healthcare specific background specific industry experience etc.)? a. Hands on experience with modern application – microservices, Cloud and CI/CD b. 5-7 years of hands on Data and AI engineering work c. Good communication with developing architecture and design documentation What skills/attributes are required (please be detailed as to number of years of experience for each skill)? • Bachelor's Degree or master's degree in Computer Science. • 5+ years of hands-on software engineering experience. • Demonstrated AI/ML solution design experience • Proven work experience in Spark, Python, SQL, Any RDBMS. • Familiarity with Azure Data Lake, Synapse, ADF, Power BI. • Experience building, deploying and maintaining ML models in production • Experience with MLOps tools such as ModelDB, MLFlow and Kubeflow. • Familiar with best practices in the data engineering and MLOps community. • Ability to convey complex concepts and ideas in a clear and concise manner to a wide range of audience internal business stakeholders, outside partners and technology teams. • To be able to work in a fast-paced agile development environment. • Proven track record in working with diverse teams to achieve goals • Strong problem solving and troubleshooting skills with the ability to exercise mature judgment. What skills/attributes are preferred (what will set a candidate apart)? • Experience with AzureML • Expert in Azure Synapse, Azure Container Registry, Azure App services Of the required skills listed, which would you consider the top 3? Please list your expectations regarding years of experience for each requirement. a. AI/ ML Solution design b. Strong problem solving and troubleshooting skills with the ability to exercise mature judgment. c. MLOps What will the interview process look like? (Video phone or in person? How many rounds? How technical will the interviews be?) a. How many rounds? 2-3 b. Video vs. phone? Video c. How technical will the interviews be? Mostly technical

To help boost women empowerment in the new normal, PLDT wireless unit Smart Communications, Inc. (Smart) has partnered with social impact start-up Connected Women for technology upskilling and livelihood opportunities for women across the Philippines. With the goal of training over a thousand women by 2021, Connected Women's Elevate AIDA (Artificial Intelligence Data Annotation) program offers online skills development and remote work opportunities in the artificial intelligence industry. Backed by UN Women, the 75,000 member-strong organization launched ConnectedWomen.ai to provide a talent pool for businesses worldwide while creating an impact for Filipino women and their families. Smart will support Connected Women's upskilling initiatives that include data labeling, remote work, professional communication, and computer skills, which are all scalable in the digital remote workspace. Participants will also benefit from career coaching, developing critical thinking and problem-solving skills and mentoring.

It's time for communication vendors to start investing in machine learning. Machine learning seems like the only topic in the computer science world today. In many cases, it comes with the terms "deep learning," or even better, "artificial intelligence" (AI). When searching for some VC funds that specialize in AI (yes, there are a few), I bumped into VC funds driven by decisions made and supported by AI. As part of a recent study we conducted on AI technologies used in real-time communications (RTC), we looked at hundreds of voice and video communications offers -- and discovered that these services are clearly lagging behind on the use of machine learning.

Over the course of the last weeks and months you couldn't escape news and stories about messaging platforms going after B2C use cases à la "order me some food", "book me a hotel room" or "I need a ride downtown in 30 minutes". Pioneered and taken to huge success in Asia by platforms like Weixin/WeChat, LINE and Kakao, especially Facebook with its two behemoth platforms Messenger and WhatsApp is taking decisive actions to bring businesses and consumers together on their platforms. Kik is even faster, having just launched such a botstore for brands. In their launch line-up are 18 well-known brands such as Sephora, H&M or The Weather Channel. And with these moves, communications platforms will tap into significant revenue streams in the form of rev shares and commissions for being the facilitator between businesses and consumers in everyday Transactions.

Microsoft wants Skype to be more than just a chat app. The company unveiled a significant Skype update that will turn it into a platform for interacting with other apps, similar to what Facebook has done with its Messenger app. This new version will allow Microsoft's virtual assistant Cortana and third-party apps to integrate with Skype. Microsoft is touting the ability to plan trips, shop, and book a hotel all within the Skype app by interacting with virtual bots. "We want all of these conversational tools to become rich canvases for computation," Microsoft CEO Satya Nadella said on stage at the company's Build conference to preface the announcement.