Future Of Work

We’ve all heard about AI’s potential to boost productivity. But what truly matters to me is whether it’s making work better for the people who show up every day. At Cisco, our People Intelligence team, in collaboration with IT, has been exploring this very topic, and the findings are fascinating. Here are five key insights from our research that leaders should take seriously: 1. Leaders are key to adoption. At Cisco, employees are 2x more likely to use AI if their direct leader uses it. 2. Generic AI training doesn’t work. Role-specific, practical training accelerates AI use. 3. Confidence gaps exist among senior leaders. Directors at Cisco often feel less confident with AI than mid-level employees, underscoring the need for tailored support at all levels. 4. Employee autonomy fuels adoption. Hybrid work environments are powerful accelerators for AI adoption, while mandates can hinder it. Employees who voluntarily go to the office are more likely to use AI, while those who are required to work on-site have lower adoption. 5. AI use is linked to employee well-being, but the relationship is complex, with both benefits and trade-offs that require thoughtful navigation. This is just the beginning. Next, we’re looking at how AI is transforming the way teams operate. For now, one thing is clear, employees who use AI aren’t just more productive. They’re also more engaged, better aligned with company strategy, and empowered to focus on meaningful work. #AIAdoption #EmployeeExperience #FutureOfWork

#Diversity in high-tech fields remains critically low. The Equal Employment Opportunity Commission (EEOC) recently reported that #Black and #Latino professionals are underrepresented in high-tech roles, especially in leadership. These numbers highlight ongoing structural barriers in hiring, promotion and retention. This gap is a missed opportunity to tap into a wealth of diverse talent and perspectives essential to the future of tech. However, addressing and thoroughly fixing these challenges will require time, consistent effort and a long-term commitment to systemic change. Companies can support the progression of representation in tech by investing in training, mentorship and internship opportunities that open doors for people who were historically shut out. Programs like internXL, a platform that is committed to increasing diversity and inclusion in the internship hiring process for top companies, are making a significant impact. Similarly, the expansion of STEM education at institutions like Cornell University is helping to connect talented young people from underrepresented communities with opportunities for high-tech careers. When we work together to remove these barriers, we’re fostering a more inclusive workforce and strengthening innovation, problem-solving and leadership in the industry. Let’s build a tech future that reflects the diversity of our society. https://bit.ly/3UNtOCh

Just out in Harvard Business Review, summary of the Hybrid Experiment results and lessons on how to make hybrid succeed. Experiment: randomize 1600 graduate employees in marketing, finance, accounting and engineering at Trip.com into 5-days a week in office, or 3-days a week in office and 2-days a week WFH. Analyzed 2 years of data. Two key results A) Hybrid and fully-in-office showed no differences in productivity, performance review grade, promotion, learning or innovation. B) Hybrid had a higher satisfaction rate, and 35% lower attrition. Quit-rate reductions were largest for female employees. Four managerial lessons 1) Hybrid needs a strong performance management system so managers don’t need to hover over employees at their desks to check their progress. Trip.com had an extensive performance review process every six months. 2) Coordinate in-office days at the team or company level. Schedule clarity prevents the frustration of coming to an empty office only to participate in Zoom calls. Trip.com coordinated WFH on Wednesday and Friday. 3) Having leadership buy-in is critical (as with most management practices). Trip.com’s CEO and C-suite all support the hybrid policy. 4) A/B test new policies (as well as products) if possible. Often new policies turn out to be unexpectedly profitable. Trip.com made millions of dollars more profits from hybrid by cutting expensive turnover.

In companies where productivity has increased by 50%, creativity has doubled, and employee satisfaction is at an all-time high, one surprising change stands out: ditching the outdated obsession with time tracking. Too many managers are stuck in an outdated paradigm, fixating on: • When employees clock in • How long they sit at their desks • Micromanaging daily schedules But we’ve hired smart, capable professionals. Treating them like children who need constant supervision is not just demeaning – it's counterproductive. However, it's crucial to maintain a balance. While micromanagement is detrimental, companies still need to ensure discipline and focus on key priorities. The goal is to empower employees while aligning their efforts with organizational objectives. That’s why one needs to focus on result-focused management: 1. Shift your metrics: Focus on project milestones, work quality, and client satisfaction instead of hours logged. 2. Embrace flexibility: Allow flexible hours and remote work when possible. Trust employees to manage their time effectively. 3. Cultivate a culture of trust: Communicate openly about priorities and challenges. Reward results, not face time. Promote work-life balance and well-being. Companies like Netflix, Basecamp, and Atlassian have implemented results-only work environments (ROWE) with remarkable success. They report higher employee engagement, better outcomes, and a more dynamic, innovative workplace culture. What's one positive outcome you've experienced (as a manager or employee) when given more autonomy at work? #Leadership #EmployeeEmpowerment #WorkplaceCulture

Last week, I described four design patterns for AI agentic workflows that I believe will drive significant progress: Reflection, Tool use, Planning and Multi-agent collaboration. Instead of having an LLM generate its final output directly, an agentic workflow prompts the LLM multiple times, giving it opportunities to build step by step to higher-quality output. Here, I'd like to discuss Reflection. It's relatively quick to implement, and I've seen it lead to surprising performance gains. You may have had the experience of prompting ChatGPT/Claude/Gemini, receiving unsatisfactory output, delivering critical feedback to help the LLM improve its response, and then getting a better response. What if you automate the step of delivering critical feedback, so the model automatically criticizes its own output and improves its response? This is the crux of Reflection. Take the task of asking an LLM to write code. We can prompt it to generate the desired code directly to carry out some task X. Then, we can prompt it to reflect on its own output, perhaps as follows: Here’s code intended for task X: [previously generated code] Check the code carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it. Sometimes this causes the LLM to spot problems and come up with constructive suggestions. Next, we can prompt the LLM with context including (i) the previously generated code and (ii) the constructive feedback, and ask it to use the feedback to rewrite the code. This can lead to a better response. Repeating the criticism/rewrite process might yield further improvements. This self-reflection process allows the LLM to spot gaps and improve its output on a variety of tasks including producing code, writing text, and answering questions. And we can go beyond self-reflection by giving the LLM tools that help evaluate its output; for example, running its code through a few unit tests to check whether it generates correct results on test cases or searching the web to double-check text output. Then it can reflect on any errors it found and come up with ideas for improvement. Further, we can implement Reflection using a multi-agent framework. I've found it convenient to create two agents, one prompted to generate good outputs and the other prompted to give constructive criticism of the first agent's output. The resulting discussion between the two agents leads to improved responses. Reflection is a relatively basic type of agentic workflow, but I've been delighted by how much it improved my applications’ results. If you’re interested in learning more about reflection, I recommend: - Self-Refine: Iterative Refinement with Self-Feedback, by Madaan et al. (2023) - Reflexion: Language Agents with Verbal Reinforcement Learning, by Shinn et al. (2023) - CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing, by Gou et al. (2024) [Original text: https://lnkd.in/g4bTuWtU ]

What if you could turn Sentinel 2’s 10m bands into 1m-resolution images across all spectral channels? That’s exactly what S2DR3 just made possible. S2DR3, a 12-band single-image super-resolution model that upscales all Sentinel-2 bands (10m, 20m, 60m) to a unified 1m/px resolution. That’s 10x enhancement on every spectral band optimized for soil and vegetation analysis, and capable of capturing spatial features as small as 3m. Some strong use cases for this include agriculture, field boundary detection, environmental monitoring, and carbon credits. These results are really impressive but as with any model there are tradeoffs. ✅ Pros Strong spectral fidelity across all 12 bands Preserves subtle soil & vegetation differences No evidence of hallucinated features (wow) Especially effective in rural and agricultural settings ❗ Cons Minor distortions in urban/structured environments Small object reconstruction (sub 5m) still limited Training bias means unusual textures underrepresented Ground truth for 12-band data is synthesized, not captured While it doesn’t solve all high resolution challenges, S2DR3 shows what is possible when spectral accuracy and spatial fidelity are treated as core design goals. Check out the Medium article in the comments. 🌎 I'm Matt and I talk about modern GIS, AI, and how geospatial is changing. 📬 Want more like this? Join 6k+ others learning from my newsletter → forrest.nyc

Smart helmets incorporating AI technology represent a significant advancement in safety, connectivity, and overall functionality. What do you think about this one? These helmets leverage artificial intelligence to enhance various aspects of the user experience. 1. Head Protection: Impact Detection: AI can be employed to detect and analyze the severity of impacts, providing real-time information about potential head injuries. Emergency Response: Smart helmets can automatically send distress signals or call for help in the event of a significant impact. 2. Augmented Reality Displays: Helmet-Mounted Displays: AR technology integrated into helmets can offer real-time information, such as navigation, speed, and relevant data, directly in the user's line of sight. Enhanced Situational Awareness: AI algorithms can analyze the surroundings and provide augmented information, like highlighting potential hazards. 3. Communication and Connectivity: Hands-Free Communication: AI-driven voice recognition enables hands-free communication, allowing users to make calls, send messages, or access information without removing the helmet. Intercom Systems: Smart helmets can facilitate communication between riders, improving group coordination during activities like motorcycling or cycling. 4. Gesture Recognition: Intuitive Controls: AI-powered gesture recognition enables users to control the features of the helmet through simple hand gestures, promoting a seamless and intuitive user experience. 5. Biometric Monitoring: Vital Signs Monitoring: Integrated biometric sensors can monitor vital signs such as heart rate and temperature, providing insights into the user's health and well-being. Fatigue Detection: AI algorithms can analyze data to detect signs of fatigue, alerting the user to take breaks when needed. 6. Navigation Assistance: Turn-by-Turn Navigation: Helmets equipped with AI can provide turn-by-turn navigation guidance, enhancing safety and convenience during travel. Route Optimization: AI algorithms can suggest optimal routes based on real-time traffic conditions. 7. Adaptive Lighting: Dynamic LED Lighting: Helmets with AI-controlled LED lighting can adapt to ambient conditions, improving visibility and safety, especially in low-light environments. 8. Automatic Tinting: Adaptive Visors: AI can control visors that automatically adjust tint based on changing light conditions, offering optimal visibility to the user. 9. Collaboration with IoT Devices: Integration with IoT: Smart helmets can seamlessly connect with other IoT devices, such as smartphones, smartwatches, or vehicle systems, creating an interconnected ecosystem. 10. Security Features: Facial Recognition: AI-driven facial recognition can enhance security by allowing authorized users access to specific features or functionalities. #innovation #helmet #ai via @ justhelmet

Louder for the people at the back 🎤 Many organisations today seem to have shifted from being institutions that develop great talent to those that primarily seek ready-made talent. This trend overlooks the immense value of individuals who, despite lacking experience, possess a great attitude, commitment, and a team-oriented mindset. These qualities often outweigh the drawbacks of hiring experienced individuals with a fixed and toxic mindset. The best organisations attract talent with their best years ahead of them, focusing on potential rather than past achievements. Let’s be clear this is more about mindset and willingness to learn and unlearn as apposed to age. To realise the incredible potential return, organisations must commit to creating an environment where continuous development is possible. This requires a multi-faceted approach: 1. Robust Training Programmes: Employers should invest in comprehensive training programmes that equip employees with the necessary skills for their roles. This includes on-the-job training, mentorship programmes, online courses, and workshops. 2. Redefining Hiring Criteria: Organisations should revise their hiring criteria to focus more on candidates’ potential and willingness to learn rather than solely on prior experience or formal qualifications. Behavioural interviews, aptitude tests, and probationary periods can help assess a candidate's ability to learn and adapt. 3. Partnerships with Educational Institutions: Companies can collaborate with educational institutions to design curricula that align with industry needs. Apprenticeship programmes, internships, and cooperative education can bridge the gap between academic learning and practical job skills. 4. Lifelong Learning Culture: Encouraging a culture of lifelong learning within organisations is crucial. Employers should provide ongoing education opportunities and support for professional development. This includes continuous skills assessment and access to resources for upskilling and reskilling. 5. Inclusive Recruitment Practices: Employers should implement inclusive recruitment practices that remove biases and barriers. Blind recruitment, diversity quotas, and targeted outreach programmes can help ensure that diverse candidates are given a fair chance. By implementing these measures, organisations can develop a workforce that is adaptable, innovative, and resilient, ensuring sustainable success and growth.

Imagine a world where technology doesn’t just support us—it redefines how we work, connect, and grow. That’s the future we’re preparing for at Kotak Mahindra Bank. Over the past month, I’ve had the incredible opportunity to lead three intensive training sessions focused on Generative AI and its game-changing potential. These weren’t just sessions—they were a bold step toward equipping our teams for the careers of tomorrow. As the pace of change accelerates, we’re not just keeping up; we’re setting the stage for what’s next. Why This Matters: The Context of Change The workplace is evolving faster than ever. A fascinating article published by The Times Of India on May 15, 2025, titled "The Hottest Jobs, 5 Yrs From Now? The Answers Will Surprise You," spotlighted industries poised for explosive growth by 2030—think Banking, AgriTech, Space Tech, and beyond. The piece predicts a surge in demand for roles like AI specialists, data scientists, and sustainability officers. This echoes the World Economic Forum Future of Jobs Report 2025, which forecasts 170 million new jobs worldwide by 2030, fueled by technology and a renewed focus on human skills like creativity and adaptability. These trends aren’t distant possibilities—they’re unfolding now. And at Kotak Mahindra Bank, we’re determined to lead the charge. Voices from the Top: What Industry Leaders Are Saying Global visionaries are sounding the alarm—and the opportunity: Elon Musk envisions a world with billions of humanoid robots and 50% of miles driven autonomously within five years. Satya Nadella and Sundar Pichai shared that AI already generates 20-30% of code at Microsoft and Google, with full automation on the horizon. Bill Gates is betting big on robotic surgery, signaling AI’s expanding role in healthcare. Reid Hoffman, the LinkedIn co-founder who predicted social media’s rise in 1997, now forecasts that traditional 9-to-5 jobs will vanish by 2034, replaced by a thriving gig economy. These statements aren’t just headlines—they’re a roadmap. The message is clear: adapt, upskill, and innovate—or risk being left behind. Looking Ahead: Your Role in the Future The next five years will be a defining chapter for careers everywhere. Whether you’re in banking, tech, or any field, the opportunity is yours to seize. Let’s make the next five years a launchpad for growth—for our careers, our communities, and our world. At Kotak, we’re proud to be pioneers in this journey, blending innovation with purpose to create a brighter tomorrow. Join us in embracing the future—it’s ours to shape. Together, let’s turn possibilities into realities. #FutureOfWork #AI #CareerDevelopment #OneKotak #BankingForTomorrow #Banker

The real challenge in AI today isn’t just building an agent—it’s scaling it reliably in production. An AI agent that works in a demo often breaks when handling large, real-world workloads. Why? Because scaling requires a layered architecture with multiple interdependent components. Here’s a breakdown of the 8 essential building blocks for scalable AI agents: 𝟭. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀 Frameworks like LangGraph (scalable task graphs), CrewAI (role-based agents), and Autogen (multi-agent workflows) provide the backbone for orchestrating complex tasks. ADK and LlamaIndex help stitch together knowledge and actions. 𝟮. 𝗧𝗼𝗼𝗹 𝗜𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 Agents don’t operate in isolation. They must plug into the real world:  • Third-party APIs for search, code, databases.  • OpenAI Functions & Tool Calling for structured execution.  • MCP (Model Context Protocol) for chaining tools consistently. 𝟯. 𝗠𝗲𝗺𝗼𝗿𝘆 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 Memory is what turns a chatbot into an evolving agent.  • Short-term memory: Zep, MemGPT.  • Long-term memory: Vector DBs (Pinecone, Weaviate), Letta.  • Hybrid memory: Combined recall + contextual reasoning.  • This ensures agents “remember” past interactions while scaling across sessions. 𝟰. 𝗥𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀 Raw LLM outputs aren’t enough. Reasoning structures enable planning and self-correction:  • ReAct (reason + act)  • Reflexion (self-feedback)  • Plan-and-Solve / Tree of Thought These frameworks help agents adapt to dynamic tasks instead of producing static responses. 𝟱. 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗕𝗮𝘀𝗲 Scalable agents need a grounding knowledge system:  • Vector DBs: Pinecone, Weaviate.  • Knowledge Graphs: Neo4j.  • Hybrid search models that blend semantic retrieval with structured reasoning. 𝟲. 𝗘𝘅𝗲𝗰𝘂𝘁𝗶𝗼𝗻 𝗘𝗻𝗴𝗶𝗻𝗲 This is the “operations layer” of an agent:  • Task control, retries, async ops.  • Latency optimization and parallel execution.  • Scaling and monitoring with platforms like Helicone. 𝟳. 𝗠𝗼𝗻𝗶𝘁𝗼𝗿𝗶𝗻𝗴 & 𝗚𝗼𝘃𝗲𝗿𝗻𝗮𝗻𝗰𝗲 No enterprise system is complete without observability:  • Langfuse, Helicone for token tracking, error monitoring, and usage analytics.  • Permissions, filters, and compliance to meet enterprise-grade requirements. 𝟴. 𝗗𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 & 𝗜𝗻𝘁𝗲𝗿𝗳𝗮𝗰𝗲𝘀 Agents must meet users where they work:  • Interfaces: Chat UI, Slack, dashboards.  • Cloud-native deployment: Docker + Kubernetes for resilience and scalability. Takeaway: Scaling AI agents is not about picking the “best LLM.” It’s about assembling the right stack of frameworks, memory, governance, and deployment pipelines—each acting as a building block in a larger system. As enterprises adopt agentic AI, the winners will be those who build with scalability in mind from day one. Question for you: When you think about scaling AI agents in your org, which area feels like the hardest gap—Memory Systems, Governance, or Execution Engines?