About
Lively and dedicated Senior Engineering Leader specialising in large-scale transformation…
Activity
6K followers
-
Dan Taylor shared thisI am particularly proud of this one - it's one of the first product features we've released that used all the brilliant new tech we've been adding to the platform. Data Lake, Streaming Events and our amazing new UI. With this kind of technology, we're going to the moon and we're bringing the world's energy system with us! 🚀 🌍 🔋Dan Taylor shared thisIn the time it took you to read this sentence, the energy market just changed. Automation in trading is often talked about as a "hands-off" solution. But handing over the reins to an automated system is a huge leap for most trading teams. To make that leap, we know you need one crucial thing, Trust. Yesterday, our CTO Roger Hollies talked about the technical AI milestones behind Trade Hub's latest release. But, I want to talk about what this actually means for the people running the desks. Our Intraday Optimisation feature, is not only "running an algorithm." The system is constantly adapting to market volatility. Literally re-optimising every single minute. What's the point of a 60-second refresh if the user does not trust what it's doing? Nothing! So we placed a heavy focus on visualisation. Traders need to see the "why" behind the decisions being taken. By showing the optimisation changing minute by minute, we give our users full insight to how their assets are responding and how the "brain" is working. No more black box trading algorithms, we are giving them a high-fidelity map and a co-pilot they can actually understand. Click through the screenshots below to see exactly how this looks in practice as the optimisation dynamically shifts and adapts minute by minute. I am so proud of the entire team for bringing this to life. This has been another one of those passion projects I have got to lead at Arenko but that only happens with the awesome team I have alongside me. #IntradayOptimisation #Nimbus #Automation #BESS #NESO #OptimiserUI
-
Dan Taylor reposted thisDan Taylor reposted thisWe are thrilled to announce that Arenko has formally expanded its partnership with RWE, doubling our supported sites to six with the addition of three new UK solar-plus-storage projects (150MW). As co-located assets become central to the energy landscape, we are proud to act as the technology partner for RWE, turning complex hybrid sites into market-ready, dispatchable units. This 150MW expansion isn't just about scale, it strengthens Arenko’s position as a leading software partner for optimising complex, grid-scale renewables and storage assets. Key areas of complexity: ⚡ DC-Coupled Innovation: We are providing the critical architecture to ensure seamless, zero-latency communication between solar, battery, and the grid. 🔋 Long-Duration Storage: Two of these new projects feature four-hour battery duration, future-proofing the systems for a fully decarbonized grid. 💻 The Digital Backbone: RWE is integrating our Nimbus platform to provide end-to-end monitoring, automated trading, and the enterprise-grade security required for utility-scale operations. The full version of the press release is available at https://lnkd.in/eaVtA-tG Read more about how we're optimizing 1.2GW+ of capacity globally: https://arenko.group/ #RenewableEnergy #BESS #SolarPower #colocation #EnergyStorage #CleanTech
-
Dan Taylor shared this
-
Dan Taylor shared thisNot gonna lie, working for Max is among the greatest perks of this role... ...but if you want to use your skills with data to Actually Change the World, that's also included. Energy is the greatest technical challenge of our time. Up for it?Dan Taylor shared thisGreat news, we've got a new role open in our Data team! Like solving genuinely interesting data problems and want to build a product that will have a real impact on the renewables industry in GB and beyond? Check it out! https://lnkd.in/eBJUnSxs
-
Dan Taylor shared thisAmazing to see the boss representing Arenko on the prestigious Modo podcast. We're so proud of our amazing tech, our exceptional platform and our wonderful people - and I think that comes across clearly listening to Roger speak. We're also deeply proud of the proactive and vocal role we play in the industry - being early adopters of new markets and technology, engaging with NESO on new policy and tech at all levels and solving tough technical challenges others shy away from. Energy is a slow moving industry - the stakes are higher here than anywhere else. The batteries and renewables we look after keep the lights on across the UK. That said, change is inevitable - as the industry moves from manual processes to software control, the way we govern and coordinate it must also change - I know for certain that Arenko's voice will help define and deliver the energy systems of the future. #energy #software #technology #engineering #dataDan Taylor shared thisIt was a long time coming but finally I made it on to the Modo Energy podcast. Thank you Ed Porter and the team (Luke Arghiros is a total pro!) for making my first ever podcast a really fun and challenging discussion. We covered a lot of ground but to give a taste of what we talked about: 🔋 The evolution of energy systems and how #BESS and #digitalisation has disrupted the energy industry and will continue to do so into a #sustainableenergy future. 🐙 How we compare to Kraken: I make a big statement here (see clip) but when it comes to managing BESS, renewables and automated trading the feedback we repeatedly get is that we are best in class. 🦸♂️ A bit on the history of Arenko and our awesome team: the reasons I will stand by statement like above all-day, everyday. 🤖 What role automation plays and how we've been using #AI for the past 7 years: the inappropriate dummer analogy..... ⬛ The optimisation black box and how we visualise decision making to gain trust (It's a thing of beauty ;)) 🦾 The challenges the global energy industry faces and my view on how we can create a pathway to over come them using tried and tested transition methods from other sectors. 🤑 How National Energy System Operator can accelerate the energy transition and ultimately reduce costs to the consumer by publishing penalty data Listen/watch now on the following platforms: Spotify - https://lnkd.in/etEq75kn Apple podcast - https://lnkd.in/eRd28MBT YouTube -https://lnkd.in/e-PKafUD Thanks again to the Modo Energy team: Keep doing what you are doing it's great. Finally: A beer/chocolate bar to anyone (from outside Arenko) who can tell me what that is on my jumper 😉
-
Dan Taylor shared thisDoes anyone else smell the whiff of desperation creeping into AI hype this week? AI is great. Unrelated to that, AI is at the heart of a Tech Industry bubble which will burst the moment markets start to believe that "infinite profits tomorrow" isn't happening. Bless 'em, the Bro-ligarchs are working flat out to delay that moment as long as they can. We've had "Terawatts", "Fusion", "Quantum" and this week Jeff decided to play the "Space" card. OpenAI is paying NVIDIA so much for hardware - that will never be built - that NVIDIA is having to fund OpenAI to keep the orders for imaginary GPUs flowing. Never mind the obvious challenges of dissipating heat in a flippin' vacuum. Ignore that you'd need 5km² of solar panels per GW of power... this is AI in SPAAAACE!!! A fusion-powered quantum AI singularity in literal flippin' orbit! I mean, talk about scope creep! I'd love to see the faces of my friends in our Product team when I tell them they can't have that new chart this week because I need to solve three or four of the most difficult problems ever conceived first. I might throw in faster than light travel, artificial gravity and the Infinite Improbability Drive too - just to see if they buy it. I mean, good grief. #ai #quantum #fusion #bubble #space #spaace #spaaaace
-
Dan Taylor shared thisThrilled to have represented Arenko at EPRI's European AI Round Table today, presenting our long track record using machine learning in the Energy Industry as part of our awesome Nimbus platform. The discussion covered a broad range of topics and left me considering a core dichotomy in the AI+Energy sphere: Is AI here to support the energy industry, or is the energy industry looking for ways to to support AI? * Data factories, Gigafactories, EU data centres * Foundational/shared models * Centralised compute, millions of GPUs It strikes me that these are all examples of energy players supporting growth in the AI space. But is that what we care about? Surely AI is a tool we should be using to support innovation and efficiency in our Energy systems? * Use-cases and real world wins - collaboration on real products and outcomes * Open data as a primary design consideration and condition of participation * Rules as code - robust but unambiguous legislation, open source, machine-readable specifications over fuzzy "legalese" I firmly believe AI can and will underpin the future energy systems of planet Earth - if we can move the focus away from compute infrastructure and onto laying solid foundations which allow real innovation (companies like Arenko the world over) to thrive. Let's go!
-
Dan Taylor shared this👹 Biases and Fallacies: The Card Game™️ (LinkedIn Edition) This week's side project is a deck of cognitive biases and logical fallacies you can collect and trade! Think Dungeons & Dragons but instead of orcs and goblins, it’s Straw Men, Slippery Slopes, and the IKEA Effect. Because these days the scary things aren't hiding in the woods or under the bed. They’re right here on your socials: “Correlation is not Causation” - waking up at 5am will not make you a CEO! “Survivorship Bias” - Elon, Mark and Jeff just got lucky, OK? “Availability Heuristics" and "Personal Incredulity" in the comments, whether you like it or not. Here’s the full set of cards: https://lnkd.in/eAeqHq57 See how many unique biases and fallacies you can find on your feed today! Credit to everyone who's posted logically questionable content here: You're the real heroes! I used AI for the pictures, then spent more time debugging AI-generated code than coding it myself would have taken.
-
Dan Taylor shared thisTech management tip of the week: the secret to productivity is repeatability. As a Software team, get good at repeating patterns: standing up an API, building an ETL, adding a new widget to your UI. Use the same solid tools, good practices and consistent style every time. Get to the point where you can stand up a walking skeleton in a day; release one or two of these things into production every sprint and trust them to work 24x7 thereafter. Conversely, if you opt for bespoke architectures, be it a quick hack or a huge new platform, you can expect to go quick today and slow tomorrow. It may seem pragmatic to do some clever tweak to save time in the short term, you're actually creating a new head scratcher for the future - when everyone's forgotten the context. Now go make that tower 20cm taller... #software #leadership #data #architecture #techDebt
Dan Taylor replied to a comment
3w
Edoardo Simioni Roger Hollies this is a really interesting debate - as a fan of both AI *and* Nimbus, let me pitch in with my two cents.
Firstly, let's be clear, this isn't "AI" in the way that 2020's LLM mania is AI. LLMs played a role in developing this - getting us to market quicker with a better product by supercharging our engineers. But LLMs are not the right tool to trade an asset.
Nimbus absolutely uses a bunch of AI to do what it does - including neural networks. NNs have been around for 80+ years and have been established as a great tool for predicting time-series for at least 25 of those. Whether it's state of health, market out-turn or the price of oil, AI is the best tool for the job.
But *AI is just a tool*. While the fever in Silicon Valley is focussed on "selling shovels", most businesses must remain focussed on solving actual problems. Nimbus solves a significant problem: using AI to automate the trading of batteries - and a day will come when all storage and generation is traded this way!
So it is 100% automation and it 100% uses AI!
What's clever about this is that we've made a huge step towards making an automated AI system *explainable* and that's huge!
Dan Taylor commented on a post
1mo
As well as Sam's comments being dystopian in the extreme - reducing human value to nothing more than productivity - it also misses a major technical difference.
LLMs learn once then stop. They take vast quantities of energy and data to reach a level, then they're done. Any new information they receive after that point lives in transient context, quickly forgotten.
Humans learn every day. Adapting and changing, testing boundaries, falling down and picking themselves up. Every experience shapes an ever changing model of the world.
Perhaps they'll crack this with AI one day, but today, Sam's comparison is meaningless.
Dan Taylor replied to a comment
1mo
Mieke Kooij honestly, it seems crazy, but this saved me more time than I'd like to admit!
-
Dan Taylor liked thisDan Taylor liked thisThis week, Trainline launched our ChatGPT app, enabling users to seamlessly search for train times. This initiative exemplifies our commitment to integrating AI across the business, following the success of our AI-powered Travel Assistant, which provides users with tools to find and book the right journey. This achievement was made possible by the talented teams in Data and Engineering who collaborated closely to drive outstanding results. Explore the app by adding it to ChatGPT and tagging @trainline.
-
Dan Taylor liked thisDan Taylor liked thisNEW UK WIND RECORD! Through the madness the #greattransition to a energy system based on renewables continues and this day highlights why a sustainable energy system is so important. On Wednesday 25th March this week the UK once again broke the record for the amount of wind generated energy on the grid. 23.8 GW from wind delivering 58% of the nations electricity over the peak demand period (interestingly over midday). See screen shot below from the National Energy System Operator app*. The entire day was dominated by wind generation (up to 80% of demand over night) with, for the most part, very little gas on the system (see second image below from the excellent Drax Group electric insights dashboard, one of the many tools to quickly look at the energy mix on the grid). Perhaps the most interesting and relevant aspect of the day however, particularly in light of the current geopolitical and macro economic back drop, is what happens to the electricity price. See the final screen shot from the Arenko Nimbus platform showing our price prediction models along side the outturn in pink. Due to wind generation generating the mix electricity prices are held low for most of the day. The moment gas is brought on to the system even at marginal levels the electricity price jumps from less than £20/MWh to over £140/MWh..... and hold there throughout the next few days to the point of this post..... While gas is still on the system. I'm not sure I have ever seen a better illustration of the issue with our addiction to imported fossil fuels and the problem with those fuels setting the price of our electricity. Get off gas. Build more renewables Build more battereis to time shift the renewable energy Collocate battereis with renewables to maximise use of the grid. *The irony in the app screen-shot is not lost: the messaging of a wonderful feat of engineering magnificence from NESO slightly ruined by poor software upkeep....
-
Dan Taylor liked thisDan Taylor liked thisLast week, we followed up on our commitment to engage with STEM outreach to mark the International Women’s Month, and invited Ann Hodgetts, Director of STEMAZING, to our office for a chat about their outreach work, mentorship programmes, the joys and challenges of working with the next generation, and what can we do to help 🤝 Over the years, quite a few Arenkians took part in initiatives to improve diversity and inclusion in STEM in general and software engineering in particular, so Ann’s talk generated a lot of interest. Ann’s lively and humorous presentation style kept us entertained as well as informed, and she even included a quiz, having heard that as a company we do enjoy a good one 🏆 Thank you, Ann! ♥️ #InternationalWomensDay #Inclusion #Diversity #WomenInSTEM
-
Dan Taylor reacted on thisDan Taylor reacted on thisKier Starmer: "What gives us control is renewables. Our own home-grown energy which is then more secure and more independent. That’s why I think we should go further and faster on renewables. Let’s get control of our own energy…so we don’t have to keep worrying that our bills are going to go up"
-
Dan Taylor reacted on thisDan Taylor reacted on this"Coding is solved" says the company whose status page looks like a Christmas sweater. Let's do a quick reality check. 1) Dario Amodei: "AI will write almost all code within 3-6 months." 2) Elon Musk: "AI will be better than any human programmer very soon." 3) Sam Altman: "We're close to AI that can do the work of a software engineer." Sounds incredible. So what happened next? Anthropic acquired Bun, because apparently the AI that's "solving coding" still needs human-built infrastructure to run on. I keep seeing servers buckle under load so regularly that the status page has more colour than a risk dashboard during a vol spike. If you're a Startup SWE or Quant Dev (the people who actually ship code under pressure). The tool is incredible. It's also unreliable, context-limited, and occasionally confidently wrong. Coding isn't "solved." We just got a very fast, very expensive, very dramatic junior dev.
-
Dan Taylor reacted on thisDan Taylor reacted on thisCome work with me! (Other job perks include “doing good” and “building cool stuff”)
-
Dan Taylor reacted on thisDan Taylor reacted on thisCome join me at Arenko, we are hiring a senior infrastructure engineer! Are you are terraform expert? Do you want to be part of the clean energy transition? Want to join an amazing team making a real difference? If so apply here.
-
Dan Taylor reacted on thisDan Taylor reacted on thisYour best trading day does not start at 8 AM. It is won or lost the day before. Last week, I shared how our new Intraday Optimisation feature gives you a real-time view into how our algorithm is adapting to the market minute by minute. Unfortunately, even if you have the best reflexes in the world, it is not good enough if you start the day with a bad game plan. Building a Day-ahead strategy is incredibly complex. Traders have to juggle so much information from weather forecasts, asset state / degradation, grid constraints, and price volatility, and then decide what is the best bet for them across a limited number of baskets. Enter our new EAC bidder as part of the Day-ahead Optimisation feature within Nimbus Trade. Now, instead of spending hours wrestling with fragmented data and complex spreadsheets, Nimbus just does the heavy computational lifting. It crunches the millions of data points required to lock in the smartest possible baseline strategy to offer to NESO. Once the auction has taken place, it schedules the awarded contracts back in Nimbus and delivers them the next day. The result? We are giving traders their time back. We are letting them step away from the calculator and get back to doing what they do best: formulating long-term trading strategies and applying their actual market intuition to steer the ship. Combining these two new features together really creates a different class of trading desk. Below I have attached a screenshot of some contracted bids won through our automation. Again, just really proud of the Arenko team for delivering both halves of this puzzle. It is completely transforming how our users interact with the market. #dayaheadOptimisation #EAC #Trading #BESS #NESO
Experience & Education
-
Arenko Group
View Dan’s full experience
See their title, tenure and more.
or
Licenses & Certifications
-
Certified Scrum Master
Scrum Alliance
Issued
Publications
-
Innate and acquired immunity in real time systems
IEEE
In most potential industrial applications of artificial immune systems for early fault detection, some form of simple fault detection system already exists. We propose that this existing layer of simple, generally rule-based fault detection is analogous to innate immunity in the natural immune system. We argue that the artificial acquired immune system should focus on the detection of fault conditions not already covered by the innate immune system, most importantly on the very early symptoms…
In most potential industrial applications of artificial immune systems for early fault detection, some form of simple fault detection system already exists. We propose that this existing layer of simple, generally rule-based fault detection is analogous to innate immunity in the natural immune system. We argue that the artificial acquired immune system should focus on the detection of fault conditions not already covered by the innate immune system, most importantly on the very early symptoms of faults which, we believe, are often very similar to self. This has implications for detector generation algorithms. We test two novel detector generation algorithms that address this issue, using temperature data from refrigerated cabinets in UK supermarkets. Results show that location of detectors in problem space is important and that detector sets concentrated close to self in problem space are better at detecting early stages of fault.
Other authors -
-
An Investigation of the Negative Selection Algorithm for Fault Detection in Refrigeration Systems
Springer
Failure of refrigerated cabinets costs millions annually to supermarkets, and a large market exists for systems which can predict such failures. Previous work, now moving towards deployment, has used neural networks to predict volumes of alarms from refrigeration system controllers, and also to predict likely refrigerant gas loss. Here, we use in-cabinet temperature data, aiming to predict faults from the pattern of temperature over time. We argue that artificial immune systems (AIS) are…
Failure of refrigerated cabinets costs millions annually to supermarkets, and a large market exists for systems which can predict such failures. Previous work, now moving towards deployment, has used neural networks to predict volumes of alarms from refrigeration system controllers, and also to predict likely refrigerant gas loss. Here, we use in-cabinet temperature data, aiming to predict faults from the pattern of temperature over time. We argue that artificial immune systems (AIS) are particularly appropriate for this, and report a series of preliminary experiments which investigate parameter and strategy choices. We also investigate a ‘differential’ encoding scheme designed to highlight essential elements of in-cabinet temperature patterns. The results prove feasibility for AIS in this application, with good self-detection rates, and a promising fault-detection rate. The best configuration of those examined seems to be that which uses the novel differential encoding with r-bits matching.
Other authors -
-
Predicting alarms in supermarket refrigeration systems using evolved neural networks and evolved rulesets
Supermarkets suffer serious financial losses owing to
problems with their refrigeration systems. Most refrigeration
units have controllers which output "high-temperature" and
similar alarms. We describe a system developed to predict
alarm volumes from this data in advance, and compare evolved
and backpropogation-trained neural networks, and evolved
rulesets for this task.Other authors -
Recommendations received
-
LinkedIn User
8 people have recommended Dan
Join now to viewOther similar profiles
-
Jim Charvill
Jim Charvill
Technical Leader specialising in Agile software development and transformation, on-premise to cloud product strategy and delivery, technical and people leadership, business partnering/outsourcing/off-shoring, M&A and technology strategy.<br><br>Coaching teams to adopt and adapt lean practices, especially in building trusted partnerships with customers and focussing on the delivering of customer value. <br><br>Experience leading software teams to deliver multiple platform products from startups to mature lifecycle, across Education, Legal, Charity, Finance, Oil & Gas, Pharma, Health & Social Care, Clinical Research & Critical Systems.
Explore more posts
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Dan Taylor in United Kingdom
1376 others named Dan Taylor in United Kingdom are on LinkedIn
See others named Dan Taylor