Lōd (@Lod_io) header
Lōd (@Lod_io) profile picture

Empowering data centers with grid-aware energy intelligence, optimized for risks, costs, and compliance.▶️ Energy Strategy ⚡️| Monitoring🔎 | Automation 🤖

Vancouver, BC
Joined December 2023
Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)Photo by Lōd (@Lod_io)
Looking for AI capacity right now? Ask every provider these six questions before you sign. #AIInfrastructure #DataCenters #AIInference
1
20
1. When is the capacity energized, and is the power firm or curtailable? 2. What Tier is the facility, and what redundancy do power and cooling have? 3. Can the site support dense AI racks, including liquid cooling? 4. What is the latency to your users, and where will your data be processed? 5. What uptime does the SLA guarantee, and what happens if it is missed? 6. What happens to your workloads if a site is curtailed or goes down?
1
12
Most providers can answer the first five. Ask them about number six. Number six is the problem we work on at LōD: keeping AI inference workloads intelligently running across distributed sites when the grid changes. Reach out if you’re weighing capacity options this year. #AIInfrastructure #DataCenters #AIInference
8
North American data center supply in primary markets has nearly doubled since 2024. Yet vacancy still fell to a record low of 1.4%. Less than 1,500 MW is available across those markets today, about six months of demand at the current pace, and 80% of what’s under construction is already preleased. (Source: CBRE, 2026) Building more is necessary. However, it won’t be fast enough on its own. Meanwhile, plenty of capacity sits in blocks of a few megawatts, or a few hundred kW. too small for one AI deployment and spread across owners, grids, and networks. We think the workload distribution the overlooked part of the AI capacity shortage, and that is what we're working towards here at LōD. More details coming soon. #AIInfrastructure #DataCenters #FlexibleCompute #Capacity
27
We keep hearing that AI capacity shortage is a supply problem. We think part of it is a workload distribution problem. BloombergNEF projects US data centers will need 194 GW by 2035. The most ever connected to the grid in one year is 7.1 GW, so even at that record pace there’s a 19 GW shortfall. McKinsey & Company expects inference to become more than half of all AI compute by 2030, and expects a shift toward smaller, distributed data centers to serve it. AI inference doesn’t need one giant campus. It can run across distributed sites, and that is the solution we’re working on at LōD. #DataCenter #AICapacity #DistributedCompute #ComputeFlexibility
1
4
152
How much does mining management software cost? It depends on who you ask (per MW, per miner, or a slice of your hashrate). We pulled 2026 public pricing and found fine print that can make a “cheap” tool expensive Getting a quote? Read the guide here: lod.io/blog/bitcoin-mining-m…
1
41
We found many Bitcoin mining platforms use identical feature names but deliver different things. "Power control" vs "demand response" and "automation" (scheduled vs real-time) are often conflated. We mapped 6 common labels across LoD, Braiins, Luxor, and Foreman. Read it all here: lod.io/blog/bitcoin-mining-m… .#BitcoinMining
37
Every mining vendor will say they're top-tier...few will point out their flaws. We did that for you: a side-by-side comparison of LōD, Braiins, Luxor, and Foreman. Honest strengths, weaknesses, and who each suits. It’s a long-term choice. Guide’s free; get it here 👇lod.io/lp/mining-platform-gu… #BitcoinMining #ERCOT
38
AI’s power problem needs more than grid expansion. Compute must become flexible. ⚡ In the latest DataCentre.ca publication, our CEO @Medi_Naseri explains LōD’s millisecond inference routing, what its Dominion Energy pilot will test, and the path to “Speed to Power.” buff.ly/LKKVUcf
1
1
90
What if the ERCOT strategy saving you money today stops paying off under the next rules? 4CP is being rewritten. 12CP is only one possible outcome—and the final model could change which peaks matter and what flexibility is worth. We mapped 3 possible futures:buff.ly/MLh4ZVZ #ERCOT #4CP #EnergyFlexibility
1
60
Picking a mining platform is a hard call, and everyone says they're the best. So we wrote an honest comparison of LōD, Braiins, Luxor, and Foreman. The real pros and cons of each, so you can decide with less guessing. Get the free guide here: lod.io/lp/mining-platform-gu… #BitcoinMining #ERCOT #Foreman #Luxor #Braiins
2
56
AI compute is becoming grid-responsive. Data Center Dynamics covered LōD’s selection into the 2026 Dominion Energy Innovation Center (DEIC) Accelerate program and our upcoming pilot with Dominion Energy. We’re building toward a future where inference workloads can flex with real-time grid conditions without giving up performance or SLAs. This is what compute flexibility looks like in practice. Read the full coverage at: datacenterdynamics.com/en/ne… #DEIC #EnergyInnovation #DominionEnergy #LōD #PJM
69
Medi Naseri will speak at the Energy Investors Forum (EIF) 2026. In "Flexible Compute: The Grid-Aware Operating Model for AI Infrastructure in a Power-Constrained World," he'll address power supply-demand challenges and introduce an innovative approach to flexibility. See y'all in Dallas on July 23. Use EIF20 for 20% at EIF2026.com. #EIF2026 #EnergyInfrastructure #BitcoinMining
1
34
LōD was selected for the 2026 Dominion Energy Innovation Center Accelerate program. We will pilot our AI inference flexibility solution with @DominionEnergy . Our technology allows data centers to respond to grid conditions while maintaining performance, compliance, and SLAs. With Virginia holding the world’s largest concentration of data center load and PJM facing rising electricity demand, LōD helps operators unlock capacity, cut energy costs, improve utilization, and scale workloads without stressing the grid. Interested in learning more? Let’s talk. #PJM #DemandResponse #DEIC #EnergyInnovation #DominionEnergy @DEIC_VA
3
57
AI is starting to operate energy infrastructure, not just analyze it. But even the smartest AI makes bad decisions with stale or poorly structured data. If you’re exploring AI for ERCOT, Bitcoin mining, or data centers, this is worth a read. 👉 lod.io/blog/ercot-real-time-…
55
Bitcoin mining margins are getting squeezed. Most operators are working hard to optimize profitability. The problem? Many are optimizing the most visible costs; not necessarily the biggest ones. If you’re chasing pennies while dollars leak elsewhere, you’re solving the wrong problem. Read more: lod.io/blog/bitcoin-mining-o… #BitcoinMining #Hashrate #HashrateOptimization
2
41
Your site looks fine. Your margins say otherwise. #BitcoinMining #Hashrate #HashrateOptimization
1
36
2.- Show ops managers what techs see, not just what the summary rolls up to. Aggregated numbers hide the outliers that cost the most. #GridAware #EnergyIntelligence #EnergyStrategy #BitcoinMining #LōD
1
10
3.- Let techs log observations across shifts. Patterns beat single alerts every time. The tightest operations aren't the ones with the best hardware. They're the ones where the tech's observation reaches the right person before it becomes a repair bill. What does your escalation path look like? #BitcoinMining #Hashrate #HashrateOptimization
9