Michael Bohnert

Hi Twitter—longtime lurker and @RANDCorporation researcher, ready to tweet. My research focuses on maritime and air systems and operations as well as industrial base management. Follow me for insights, facts, and analysis of these and other topics.

@sudoingX Absolutely great piece. Fully agree. This is a critical point that isn't foot stomped enough

i get called out a lot for how long my build is taking, when someone can vibe-code an app in a weekend. anon the honest answer going to sting some of you. i'm not just building. i'm learning to tell good code from bad code. give the same prompt to five frontier models and you get five different answers, and if you can't tell which one is actually good, you're not building software, you're stacking a tower of stuff you don't understand. it runs great right up until it breaks, and then you're standing in a codebase you can't read with no idea what to touch. that's the part the "you don't need to learn to code" crowd never mentions. and notice who's pushing that line, it's the people selling you the thing that replaces learning. follow the incentive, you don't even need a conspiracy. vibe-coding is genuinely great until the day the model can't fix the bug and neither can you. that's the moment you find out you're in the middle of the ocean and you never learned to swim. the fundamentals, the unit tests, knowing WHY the code is shaped the way it is, that's your way back to shore. nobody's swimming out to get you. so here's the real take, this is the best time in history to learn to code, not the excuse to skip it. these models make you faster at what you understand and dangerous at what you don't. learn the fundamentals, write the tests, actually understand the thing. be a better engineer, not a faster button presser. i'll take the slower build i can fix at 3am over the weekend demo that falls apart the first time it touches reality.

@wndlw8 Great point.

@mbohnert Also Ukraine drones can now fly in wet weather, which Moscow has had a lot of. Simple rain can disrupt the signals.

Now that the oil and ash have somewhat settled from this week's Ukrainian strikes on Moscow, I want to make a few updates. Earlier I had commented on the quality of Russian air defenses, there is much to unpack. What was surprising was how high some of the Ukrainian drones and missiles flew. Many were flying what appeara to be between 100 and 300 feet. At this altitude radar visibility should have been in the dozens of miles, even in a city with buildings. This leads to my first observation, many of the Russian fire units appear to be unready for the incoming drones. Many of the shots are inly after the drone has passed. Either the operators are untrained (highly likely), the command and control is poor (I have yet to hear radios in the background of fire teams shot videos), or there is next to no early warning. The lack of early warning is of particular interest. Ukraine has spent significant effort targeting Russian air defenses over the past +4 years. Russia may have lost many more radars than we realized or they are intentionally keeping them off to protect them. Very hard to tell at this time, but worth monitoring. The command and control is very hard to accurately assess remotely, but there does seem to be a lack of planning and coordination of air defenses. Numerous videos show Russia air defense teams in odd locations such as low points. Now these are near roads, so it could be a choice for mobile air defense or to make up for a lack of capacity. However, having air defense units at highr elevation would support better early warning by the local air defense if broader early warning is unavailable. It is very hard to separate poor training from poor equipment. The image I attached has become widely viewed. It shows a Russian air defender shoot a man portable air defense (MANPAD). The missile immediately veers off somewhere else. It could be operator error either with poor shot selection or general procedures. It could be faulty systems providing poor feedback. It could be older systems that can't handle saturation well and detected a larger heat source offbore (noted by @TrentTelenko) . The missile that cause the iconic fuel tank discussion was a stray missile where the intended target isn't visible. More images of these failures keep pouring in. Regardless of the reason, there is something fundamentally wrong with Russian air defense. It wasn't just Ukraine's use of overwhelming raids, good route planning, and long term preparations. We may have to wait for the next bug strikes to get more information. We probably won't have to wait long. Image source @Osinttechnical

@me_too @IlvesToomas Great question. It is surprising so much got through, especially given the crackdown on vpns and Telegram. It could be that the alterstives to Max that Russians are uaing are a little better at getting around the firewalls.

@mbohnert @IlvesToomas I'm even more surprised we got to see all these videos of destruction, compared to Israel. Is there no state control and suppression in Russia anymore? Free internet for everyone? Is Putin losing the reins?

@TrentTelenko I think Russia is inflating the overall numbers. Ukraine has no reason to correct them.

Saturation of a missile based air defense is simple arithmetic. The number of ready fire missiles is know as is the reload rate. You just send enough more drones and missiles than the calculated interceptor number to destroy your target. In 2026, Ukraine has enough drones to do this.

@MalcontentmentT I mean probably one of the less surprising developments of the past few years.

I did not have a Spanish company developing a laser-guidance kit for legacy 122mm Grad rockets that improves the CEP to 3 meters, on my BINGO card.

@zubapita Well, fancy tech helps, but yes planning and preparation definitely take you very far when planning strikes.

敵の防空ミサイル網をかいくぐるのに必要なのは、映画みたいな気合いと根性と運命の女神じゃなくて、綿密な調査と計画と飽和攻撃なのですね。そりゃそうだ。

@Arthur_Foxache @Oglokoog Well, given how many missiles use the same launch setup, hard to tell which one. But yes, something is clearly amiss. That being said, Pansirs underperformed in Libya and elsewhere.

@mbohnert @Oglokoog If so, that makes it even more comical. Aren't the 57E6 radar-guided? Meaning they locked on to the oil tank, or the missile's components are so poor it went completely stupid on launch.

@morganlinton @elonmusk @nikitabier I appreciate the freedorm nature. It has definitely been nice using X to interact with new people. Especially helpful to keep share knowledge about rapidly changing AI topics.

For years, I was hesitant to post a lot on X, afraid of what people might think if I was just totally myself, and said what was on my mind. But I have to say, with @elonmusk buying X, and adding @nikitabier to the team, it is the best it has ever been. And I think this has empowered me, and many others, to feel comfortable, just being ourselves, in public, and it feels really good. Thanks to both of you for making X a place where quirky people like me, can find other quirky people like me, that seem to get excited about the same weird stuff. There is no close second to X right now, it is the place to find your people.

@Oglokoog @Arthur_Foxache This wasn't the gas tank one. That was most likely a Pansir. Similar failure mode though.

@Arthur_Foxache @mbohnert We don't know if those were the same missile.

@RichardfromSyd1 One thing to think about is because Ukraine was being struck from day one, they burned through their legacy stuff quickly and have been iterating since. Russia really didn't have to increase air defense investments until recently. They clearly failed to plan.

@mbohnert At least initially, Ukraine was using the same or older kit, yet don’t seems to have had the same issues or resolved them quickly.

@Arthur_Foxache You and me both. A few of the others may make sense as they might veer off to other heat sensors if the sensor was dirty. But yes, it is baffling.

@mbohnert What I don't understand is how an IR head MANPAD can go so stupid it ignores the target and instead locks on to a giant oil storage tank.

@CarolineBajada @clement_molin Well, I guess I will be busy tomorrow

@mbohnert @clement_molin I have to post this…

We have so much videos of Ukrainian 🇺🇦 middle-strikes now that Magyar is posting nearly one per day, not counting other units. We can easily estimate that more than 100 mid-range strikes are happening everyday in Russia's rear with at least 10 trucks hit a day (minimum)...

@BonnefonO These past two weeks in the war definitely add to that impression

@mbohnert I just finished reading your analysis. Either this is incompetence on a staggering scale or evidence of a deeper institutional breakdown. My time in Russia left me with the impression of a country so brutal and so thoroughly corrupted that its institutions barely function anymore

Hundreds of Ukrainian drones overwhelmed Moscow’s layered air defenses on June 17-18. Russia claims it downed 555+ UAVs that night alone, yet multiple struck the Kapotnya oil refinery for the second time erupting massive fires. This wasn’t luck. It was engineered saturation. The math tells the story. Ukraine launched massive coordinated waves, mixing armed drones with decoys that carry adjustable radar signatures to mimic real threats. Russian systems, optimized for fewer high-value missiles or aircraft, faced dozens arriving simultaneously from different angles. Even a 90%+ interception rate (per some Russian milbloggers) still left enough penetrators to hit the target. This breakthrough built on years of preparation. Ukrainian forces systematically targeted Russian air defense radars, launchers, and electronic warfare modules across multiple sectors. By hitting dispersed sites and forcing Russia to spread its best systems thin, Kyiv created exploitable gaps even around the heavily defended capital region. The provided inage shows how air defense locations are roughly 5 to 10 miles apart. For reference, an air defense can see drones at 100 ft about 5 miles away due to the radar horizon. This leaves at most three overlapping systems over a single target with roughly 4-32 interceptors available for use, easily overwhelmed. Flight paths were meticulously planned — sometimes J-shaped routes that circled defenses before final approach. Advanced Ukrainian drones, including faster jet-powered variants alongside propeller-driven models, complicated tracking. Decoys drew fire while real munitions pressed through. Going back to our map, buildings can be used as cover, although one drone hit a crane while flying low. The result: visible “oil rain,” refinery fires, and all four major Moscow airports grounding flights. Specific damage at Kapotnya included the combined oil refining unit, a secondary processing unit, and tank farm. Moscow Mayor Sergei Sobyanin acknowledged strikes on the refinery and minor damage elsewhere. Russia’s Defense Ministry reported broad intercepts, but geolocated footage and official admissions confirmed real effects deep inside the capital’s defensive bubble. Compare the scale: Ukraine fired over 1,000 drones toward Russia in the broader operation that day. Even partial success against the most protected targets demonstrates a repeatable capability. Low-cost mass + prior attrition of defender systems + adaptive routing = a template that exposes vulnerabilities in legacy air defense doctrine. These aren’t one-off raids. They’re the product of sustained tactical evolution. Ukraine is proving it can impose real costs on Russia’s rear areas despite heavy defenses. Full details and context in the original report. Image via @Telegraph

@CarolineBajada Thank you.

@mbohnert The team really is dedicated I’ve been listening to them since day one Their principles, their moral compass, all their work & words are brilliant! It’ll be great to listen to you again, next time you’re on… In the meantime—keep up your amazing work! Truly well done🇺🇦

And I am so glad you liked it. The team there is really dedicated and great to work with. Not sure when I will be on next. We always plan it when a new technology or capability affects the war. The Ukrainian logistics strikes with "AI" and the talk of AI with air defenses was this catalyst

@mbohnert 😁 I’m sure! Hope your flight was good and you arrived back safe and well Thank you again for coming on Ukraine: the latest It was great listening to all your very interesting insights Hope to hear you again very soon! 🇺🇦

@CarolineBajada My posts got me through the first 30 minutes 😀

@mbohnert Quite a long time then!!🙂🙂

@CarolineBajada Thank you. It helped I had a 9 hour flight.

@mbohnert 1—may we not wait long at all! 2—a great analysis! What patience and knowledge to write all this! 💪🇺🇦💪

@BonnefonO Russia should have been. They have had 2 years of strikes. There is something structurally wrong with their system. It is worthwhile to understand why it is so brittle.

@mbohnert Outstanding analysis. Thank you, @mbohnert. Ukraine has pioneered a new operational model that is reshaping the battlefield and leveling the playing field. Russia wasn’t ready for it. Europe and the United States probably aren’t either.

@BohunkPundit @Deadlyey5 Lots of webcams and windows.

@Deadlyey5 @mbohnert Ukraine can probably get this information through humint.

@Deadlyey5 Given the terrain, peoplex and satellites , probably not hard to keep up to date with minimal effort

@mbohnert It also seems like western intelligence has some program that effectively provides Ukraine with a Real time map of air defense systems and coverage so Ukraine go through gaps or oversaturate less defended areas