Been digging deeper into LangGraph while working on a few consultancy projects, past the basic demo where an agent calls a tool once and answers.
Once agents actually go live, new problems show up. One agent trying to do everything gets confused, so it helps to split the work into smaller agents, each handling one job, with a manager agent deciding who does what. Memory also needs to be saved properly, so the agent picks up where it left off instead of forgetting everything after a restart. And once multiple agents are working together, understanding what went wrong takes more than a print statement, it needs proper logs showing what each agent did and in what order.
One thing from my own work with edge devices, small hardware, limited compute: run a quick check locally first, and only send the tricky cases to a bigger model. Saves time and cost, and it’s a simple idea once you see it.
Can you measure ignorance? Most people use AI to get answers. But real-world decision making is not just about answers. It is about managing uncertainties. Check out my new blog where I discuss Bayesian Optimization through the lens of finding oil in an unknown location, where guessing becomes expensive.
In an industrial environment, it is common practice to restrict access to sensitive equipment and machinery to authorized personnel only. To enhance security and minimize the risk of unauthorized access or tampering, smart locks can be integrated into industrial panels, providing tight control and monitoring of access.
Smart locks represent a prime example of how embedded systems can be utilized to create secure and convenient access control solutions.
In this video, I will be presenting information on smart locks, its various applications and a demonstration of their proof of concept.
Hope you have subscribed to my new Youtube channel TinkerTechKnow.
Here I will be posting interesting videos and discuss about different opportunities in industry. The upcoming video is regarding the untapped IoT opportunities in foundries.
I’m pleased to announce that I’m launching my own YouTube channel, “Tinker Tech Know” where I’ll be sharing my knowledge and expertise on various topics related to technology solutions.
Over the years, I’ve had the privilege of working with clients who presented unique problem statements, which has further fuelled my passion for researching and developing innovative solutions.
Through my blog, training sessions, and internships, I’ve been sharing my knowledge and expertise, and I’m thrilled to expand my reach through YouTube. In my first video, I’ll be discussing our bus bar temperature monitoring solution in detail, which was developed to address a specific problem statement.
I hope you’ll join me on this journey and find my videos informative and helpful.
So, don’t forget to like, save, subscribe, and share the video! Stay tuned for more exciting content on my channel.
Sudden rise of IoT and allied technology has shown a good promise to BLE enabled devices and its market! BLE enabled devices are the devices that can be controlled remotely from your smartphone or tablet by connecting them with Bluetooth with minimum power consumption. Hence the name BLE (Bluetooth low energy).
“FitBit” is a popular brand which is based on BLE technology. The fitness bands and watches tracks all track steps. The accelerometer takes the movement data and translates it into digital measurements, which it how Fitbits count your steps, and measure the distance you’ve traveled, calories burned, and sleep quality. Isn’t it cool? This might create an interest to explore and understand more about BLE and its technology!
What is BLE?
Bluetooth Low Energy hit the market in 2011 as Bluetooth 4.0. When talking about Bluetooth Low Energy vs. Bluetooth, the key difference is in Bluetooth 4.0’s low power consumption. With Bluetooth LE’s power consumption, applications can run on a small battery for four to five years. Although this isn’t ideal for talking on the phone, it is vital for applications that only need to exchange small amounts of data periodically.
Just like Bluetooth, BLE operates in the 2.4 GHz ISM band. Unlike classic Bluetooth, however, BLE remains in sleep mode constantly except for when a connection is initiated. The actual connection times are only a few milliseconds, unlike Bluetooth which would take ~100 milliseconds. The reason the connections are so short, is that the data rates are so high at 1 Mb/s.
BLE’s M2M/IoT Applications:
Blood pressure monitors
Fibit-like devices
Industrial monitoring sensors
Geography-based, targeted promotions (iBeacon)
Public transportation apps
Asset tracking in industries
Locks
Smart Locks – Keyless entry!
Gone are the days where you had to worry about your lost keys and asset security. Smart Lock is an evolved lock wherein you don’t need keys or those cumbersome combinations to open it!
A smart entry system is an electronic lock that controls access to a building or door without using a traditional mechanical key.
Smart Lock uses a key which is cryptographic key that is operated from your android mobile phone. It is a secured way of using locks, thanks to BLE technology. One may ask, why the BLE locks are better than the traditional locks? Let me explain.
Highly secure: Secure standards and predefined protocols that provide maximum security digitally are used along with the physical strength and durability of the product!
Battery Life & Backup : low-battery notification that constantly notifies you of your battery status
Remote Management through App
Bi-Embedly has recently developed a solution that caters to this problem and designed a Smart Lock that can operate on both Wi-Fi and BLE technology.
A video showing its working on a padlock.
Every Lock has a secure and encrypted password which can be accessed by authorized login through mobile phone.
Panel manufacturers, Industrial sector for these locks may contact on
info@bi-embedly.com
Don’t forget to like, comment,follow and share my blog! Keep reading!
This is the recent development of Bi-Embedly Technologies LLP.
A pressure alarm system for an industrial purpose. The device measures pressure from the pressure sensor within a range of 0-400 bar. It gives a standard 4-20mA output. This pressure is checked against the set-point set using the features on the panel unit. If the actual measured pressure drops below the set-point pressure value, the alarm is raised. It is a robust and accurate system.
A smoke alarm is critical for the early detection of a fire
in your home and could mean the difference between life and death. Fires can
occur in a variety of ways and in any room of your home. But no matter where or
how, having a smoke alarm is the first key step toward your family’s safety.
Smoke detectors or smoke alarms are devices that can detect smoke in the air and therefore alert against any potential fire.
Every other day, we read news about some fire incident and deaths due to smoke suffocation etc. Here’s a news I read in yesterday’s newspaper.
Every year, millions of fires occur worldwide and a large proportion of these happen in homes. Many injuries and fatalities are also caused by these home fires, which could have been minimized had many of these homes installed smoke detectors. The difference between life and death can often be just a few minutes, which is what smoke detectors are designed for. It is estimated that smoke detectors reduce the fatality risk by 50%.
Here at Bi-Embedly we are aiming at a solution which will alert the user of a potential fire and smoke in the home/other premises. A Wi-Fi based smoke detector that detects smoke, sends data on a cloud, can also send notifications in form of messages to relatives and ambulance/Fire Brigade nearby to the smoke detector location. Here is a close look at the product.
Steps to connect and use:
Power on the smoke detector. Initial beep
indicates it’s turned on.
Open network options in your laptop or desktop.
Smoke_Detector network should be visible.
Wi-Fi settings and HTTP settings for server end-point are configurable
After the configuration, Open the cloud end-point dashboard and you will be able to see the device online and the status from the sensor. Sensor output in ppm and also the status of the alarm.
No Smoke detected, Smoke detector is online
After exposing the sensor to smoke
Smoke is detected along with the PPM readings
Using the cloud potential, you can also add some custom rules and logic for the optimum use and management of cluster of such sensors.
