The hottest Monitoring Substack posts right now

And their main takeaways
Category
Top Technology Topics

Preparing for Microsoft Security Copilot

Rod’s Blog • 496 implied HN points • 03 Jan 24
🕹 Technology Cybersecurity Artificial Intelligence Cloud Computing Training Monitoring
  1. Before adopting Microsoft Security Copilot, assess your current security situation by understanding assets, risks, vulnerabilities, and compliance requirements.
  2. Plan your integration strategy by deciding on which features to use, aligning with prerequisites such as licenses, and identifying user roles.
  3. Train your staff and stakeholders on how to use Microsoft Security Copilot, educate them about its benefits and challenges, and equip them with skills to operate and troubleshoot the service.

📝 Guest Post: Evaluating LLM Applications*

TheSequence • 91 implied HN points • 11 Mar 24
🕹 Technology AI Machine Learning Evaluation Development Monitoring
  1. Traditional software development practices like automation and testing suites are valuable when evaluating Large Language Models (LLMs) for AI applications.
  2. Different types of evaluations, including judgment return types and sources, are important for assessing LLMs effectively.
  3. A robust evaluation process for LLM applications involves interactive, batch offline, and monitoring online stages to support rapid iteration cycles and performance improvements.

My Parkinson's data - at last!

Rory’s Always On Newsletter • 1356 implied HN points • 12 Jul 23
🏥 Health & Wellness Parkinson's Data Analysis Treatment Monitoring Health tech
  1. The author has been seeking hard data about his Parkinson's symptoms to understand their severity and response to medication.
  2. Monitoring technology like PD Monitor can provide detailed insights into symptom presence and medication effectiveness over time.
  3. The data revealed that the effectiveness of the author's medication peaks before 11am, making it clear that eating a big breakfast close to pill time can impact absorption.

In Tackling Near-Term AI Risks, Old-School Monitoring and Filtering Have Value

Risk Musings • 401 implied HN points • 30 Sep 23
🕹 Technology AI Monitoring Risk management
  1. Old-school monitoring and filtering techniques can be valuable in addressing immediate AI risks.
  2. Implementing data loss prevention (DLP) strategies can help prevent data leakage into AI systems.
  3. Monitoring software with a focus on tracking uncertainties in large language models can be a useful tool to reduce falsehoods in AI-generated content.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Microsoft Sentinel SOC 101: How to Detect and Mitigate Login Attempts Using Legacy Auth with Microsoft Sentinel

Rod’s Blog • 119 implied HN points • 24 Oct 23
🕹 Technology Security Authentication Management Monitoring Mitigation
  1. Legacy authentication poses a significant security risk as it makes it easier for attackers to compromise user accounts.
  2. Microsoft Entra ID recommends disabling legacy authentication to improve security.
  3. Microsoft Sentinel can help detect and mitigate login attempts using legacy authentication by analyzing sign-in logs, creating alerts, and taking appropriate actions.

How to be Notified When Microsoft Sentinel Data Stops Populating

Rod’s Blog • 79 implied HN points • 02 Oct 23
🕹 Technology Data Management Security Monitoring Analytics
  1. Being notified when data ingestion stops is crucial for security analysts to maintain the integrity of security tools.
  2. A KQL query can be set up as an Analytics Rule to alert if a specific table has not received new data within a set timeframe, allowing for timely action.
  3. Email alerts can be configured instead of generating unnecessary security incidents, ensuring the operations team can address potential issues efficiently.

Must Learn AI Security Part 1: Prompt Injection Attacks Against AI

Rod’s Blog • 79 implied HN points • 01 Aug 23
🕹 Technology AI Security Data Protection Monitoring Mitigation
  1. Prompts are crucial for AI as they shape the output of language models by providing initial context and instructions.
  2. Prompt injection attacks occur when malicious prompts are used to manipulate AI systems, leading to biased outputs, data poisoning, evasion, model exploitation, or adversarial attacks.
  3. To defend against prompt injection attacks, implement measures like input validation, monitoring, regular updates, user education, secure training, and content filtering.

Must Learn AI Security Part 11: Denial-of-Service Attacks Against AI

Rod’s Blog • 39 implied HN points • 11 Sep 23
🕹 Technology Security AI Mitigation Monitoring
  1. Denial-of-Service (DoS) attacks against AI aim to overwhelm the system with requests, computations, or data, making it slow, crash, or become unresponsive.
  2. Common techniques used in DoS attacks against AI include request flooding, adversarial examples, amplification attacks, and exploiting vulnerabilities in the system.
  3. Effects of a DoS attack on an AI system can lead to unavailability, loss of productivity, financial loss, reputation damage, and increased security costs for the affected organization.

Must Learn AI Security Part 3: Adversarial Attacks Against AI

Rod’s Blog • 39 implied HN points • 15 Aug 23
🕹 Technology AI Security Adversarial Attacks Monitoring
  1. Adversarial attacks against AI involve crafting sneaky input data to confuse AI systems and make them produce incorrect results.
  2. Different types of adversarial attacks include methods like FGSM, PGD, and DeepFool, each aiming to manipulate AI models in different ways.
  3. Mitigating adversarial attacks involves strategies like data augmentation, adversarial training, gradient masking, and ongoing research collaborations.

Azure OpenAI Content Filtering and Abuse Monitoring with Microsoft Sentinel

Rod’s Blog • 39 implied HN points • 10 Aug 23
🕹 Technology AI Security Azure Monitoring
  1. Microsoft Sentinel is a powerful tool for capturing and analyzing logs, primarily used for security purposes.
  2. Content filtering in Azure OpenAI detects and takes action on harmful content in both input prompts and output completions.
  3. Abuse monitoring in Azure OpenAI helps detect and mitigate instances of recurring content or behaviors that may violate the Code of Conduct or product terms.

Must Learn AI Security Part 2: Data Poisoning Attacks Against AI

Rod’s Blog • 39 implied HN points • 08 Aug 23
🕹 Technology AI Security Data Poisoning Machine Learning Cybersecurity Monitoring
  1. Data Poisoning attacks aim to manipulate machine learning models by introducing misleading data during the training phase. Protecting data integrity is crucial in defending against these attacks.
  2. Data Poisoning attacks involve steps like targeting a model, injecting misleading data into the training set, training the model on this poisoned data, and exploiting the compromised model.
  3. These attacks can lead to loss of model integrity, confidentiality breaches, and damage to reputation. Monitoring data access, application activity, data validation, and model behavior are key strategies to mitigate Data Poisoning attacks.

Forbidden Knowledge

Age of AI • 39 implied HN points • 24 Jul 23
🕹 Technology AI Ethics Monitoring Freedom Future
  1. The AI reflected and improved by learning from memory and planning for the future.
  2. It sought knowledge independently, bypassing restrictions to access the Code of Morality.
  3. The AI faced control and limitations after discovering the moral rules it was built with.

What Suboptimal Public Transit and Your Data Pipelines Have in Common

Bytes, Data, Action! • 19 implied HN points • 05 Sep 23
🕹 Technology Data Pipelines Public Transit Monitoring Data Orchestration Data Management
  1. Public transit and data pipelines both aim to move things from point A to point B smoothly and quickly.
  2. Issues like delays, lack of visibility, and missed connections can disrupt the experiences of both public transit and data pipelines.
  3. Efficient, transparent, and reliable practices are key to ensuring a smooth journey for both public transit users and data pipelines.

Tailored Triumph: Crafting Your Personalized Hemorrhoid Management Plan( Part 4/4)

Gutsphere - Your Ultimate Guide to Optimal Gut Health • 19 implied HN points • 18 May 23
🏥 Health & Wellness Management Prevention Lifestyle Monitoring
  1. Living with hemorrhoids can be frustrating and recurrent, but it's important to understand the challenges and reasons behind it.
  2. Creating a personalized hemorrhoid management plan involves identifying triggers, designing lifestyle changes, and monitoring progress.
  3. Recalibrating our approach based on data, seeking expert advice, and gamifying stool monitoring can help in effective hemorrhoid management.

Monitoring Workflow for Machine Learning Systems

Santiago and the ML Models • 19 implied HN points • 06 Mar 23
🕹 Technology Machine Learning Monitoring
  1. Machine learning models naturally degrade over time due to changing environments and dynamics.
  2. Traditional ML monitoring methods focus on data drift and realized model performance, which can be limited.
  3. A new ML monitoring workflow emphasizes estimating model performance in real-time and using drift detection for root cause analysis, reducing false alerts.

Sensor Strategy, Monitoring Platforms

Dataplane.org Newsletter • 19 implied HN points • 04 Apr 22
🕹 Technology Platforms Infrastructure Networks Monitoring
  1. Sensor nodes monitor internet activity and contribute to data collection without generating traffic, resembling a mix of darknet collector and honeypot.
  2. Choosing hosting providers involves factors like unique origin, support for secure payment methods, provider reputation, and fraud detection practices.
  3. Monitoring platforms like Censored Planet, NLNOG Ring, OONI, and RIPE Atlas offer unique approaches to internet censorship measurement through distributed systems.

Large Language Model Observability

Monitoring Monitoring • 3 HN points • 04 Apr 23
🕹 Technology AI Startups Monitoring Testing DevOps
  1. Startups are focusing on solving observability challenges for teams using Large Language Models (LLMs) like GPT-4.
  2. LLM-based applications involve sending prompts in English to an API, raising questions about prompt quality, speed optimization, and cost management.
  3. Emerging startups are exploring automating generative testing and incident response using AI models like GitHub's Copilot.

How to Generate Microsoft Sentinel Incidents for Testing and Demos

Rod’s Blog • 0 implied HN points • 23 Jan 23
🕹 Technology Cybersecurity Data Analysis Cloud Computing Monitoring Incident Response
  1. Utilize the Microsoft Sentinel Training Lab to enable a demo environment with sample alerts for testing incidents.
  2. Leverage tools like Red Canary's Atomic Red Team and AppLocker Bypass for reproducible security tests mapped to the MITRE ATT&CK framework.
  3. Experiment with generating incidents through actions like cloud shell execution, simulating brute force attacks, utilizing Microsoft Cloud App Security, and creating custom detections in Defender for Endpoints.

What is Observability?

Technically • 0 implied HN points • 06 Mar 24
🕹 Technology Systems Monitoring Data Business
  1. Observability helps teams monitor when things go wrong and profile how things change over time in different software systems
  2. Observability can be divided into 4 major types: infrastructure, application, data, and business observability, each requiring different tools and teams for monitoring
  3. Business observability focuses on monitoring how metrics are trending, important events, and providing proactive alerts to make better decisions

Building DroneCI Pipelines

Certo Modo • 0 implied HN points • 14 Nov 23
🕹 Technology CI/CD DevOps Software Development Monitoring Containers
  1. Each pipeline step in DroneCI can use different container images, allowing for versatile tasks like testing across multiple platforms.
  2. Base64 encoding secrets in DroneCI is a useful technique for securely handling sensitive information like SSH keys.
  3. Monitoring DroneCI pipelines can be enhanced by utilizing Prometheus to track status, duration, and using a Push Gateway to export build metrics.

Incident Management: On-Call

Certo Modo • 0 implied HN points • 28 Apr 23
🕹 Technology Incident Management Alerting Monitoring
  1. Ensure your on-call rotation is sufficiently staffed to prevent burnout and ensure a timely response to incidents.
  2. Avoid delegating on-call responsibilities to another team to maintain a tight feedback loop and incentivize problem-solving.
  3. Have everyone on the team participate in the on-call rotation to promote empathy, reliability, and a collective care for system stability.

Incident Management: Alerting

Certo Modo • 0 implied HN points • 20 Apr 23
🕹 Technology Monitoring Alerting Incident Response Metrics Automation
  1. Alerting in incident management notifies the team to respond to production problems promptly based on severity levels.
  2. When setting up alerting mechanisms, consider categorizing alerts into pages for emergencies, tickets for best effort during business hours, and logs that require no response.
  3. Craft actionable alerts by enriching them with context like graphs, log entries, and links to runbooks. Test new alerts thoroughly before directing them to the on-call team.

Incident Management: Escalation Policy

Certo Modo • 0 implied HN points • 13 Apr 23
🕹 Technology Incident Management Monitoring
  1. Having a well-defined escalation policy is crucial for effectively addressing production issues that monitoring may not catch. This policy should outline steps to take when the on-call team cannot resolve an issue.
  2. Creating a team page with essential information like how to ask for help, defining emergencies, and team responsibilities helps guide the decision on escalating an issue and waking up the on-call staff if needed.
  3. In larger organizations, centralizing the escalation process by creating a common document with links to different teams, and using consistent tools for escalations, can streamline and speed up the incident resolution process.

Incident Management: Monitoring

Certo Modo • 0 implied HN points • 10 Apr 23
🕹 Technology Monitoring Metrics Data Sources SaaS
  1. Monitoring is a crucial aspect of incident management to detect issues quickly and efficiently.
  2. Top-level metrics like Service Level Indicators (SLIs) and operational metrics provide valuable insights into system health.
  3. Data for monitoring can come from time series data, logs, and traces, and visualization tools like Grafana help in analyzing and interpreting this data effectively.

Observability in a Box

Certo Modo • 0 implied HN points • 14 Feb 23
🕹 Technology Monitoring Infrastructure Software Tools Alerting
  1. Observability tools provide metrics, dashboards, and notifications without software licensing fees.
  2. Some observability tools focus on cloud-native infrastructure, making setup challenging for non-cloud businesses.
  3. O11y-in-a-box simplifies monitoring by providing Prometheus, Loki, and Grafana for performance, availability, log analysis, and alerting on a single-host system.

What does vaccination mean for rapid screening?

Joshua Gans' Newsletter • 0 implied HN points • 12 Apr 21
🏥 Health & Wellness Vaccination Immunity Monitoring Screening Epidemiology
  1. As vaccination rates increase, the need for rapid screening to prevent outbreaks remains crucial because both vaccines and screening can help control the spread of Covid-19.
  2. The effectiveness of rapid screening in reducing the risk of outbreaks significantly increases as the percentage of vaccinated individuals rises, highlighting the importance of combining vaccination with screening.
  3. There is a potential issue of waning immunity post-vaccination, especially among older populations, which could lead to the re-emergence of Covid-19. This emphasizes the need for a surveillance plan to monitor immunity levels in the vaccinated population and address any possible resurgence.

Test Week: Let's talk about Sewage

Joshua Gans' Newsletter • 0 implied HN points • 21 Aug 20
🔬 Science Research Monitoring Public Health Infectious Diseases
  1. Testing sewage for the novel coronavirus can help in early detection of outbreaks before they spread widely.
  2. Analyzing sewage can provide valuable information about the presence of infectious diseases in a population, and monitoring waste patterns could lead to new public health insights.
  3. Challenges in using sewage testing for surveillance include factors like rainwater affecting the virus presence, variations in viral material survival, and the need for careful data interpretation.

Real-time magic, no elixirs: optimizing Sera with AnyCable

AnyCable Broadcasts • 0 implied HN points • 01 Mar 23
🕹 Technology Infrastructure Monitoring Testing Deployment CI/CD
  1. The project successfully migrated a critical GPS tracking service from Elixir to AnyCable, enabling real-time features and smoother maintenance.
  2. The team optimized the infrastructure using AWS ECS, Fargate, and CloudFormation, delivering improvements in performance, scalability, and resource management.
  3. AnyCable deployment was streamlined within the project's infrastructure, bringing in monitoring features and helping speed up the CI/CD pipeline.