Two tracks in parallel across my career: building software and shaping policy. Started in embedded systems and DevOps (CI/CD, cloud-native infra, security auditing), crossed into government - advised the Mayor of Bucharest on a city-wide digital parking overhaul and served on the executive board of the municipal parking company.
Now back to building: working part-time at an EdTech startup applying the modern AI stack (Python, TypeScript, LLM APIs, RAG, agentic workflows) to real products on GCP.
Engineering depth + public-sector leadership + hands-on AI. Looking for a role where technical ability and strategic thinking both matter - ideally at the intersection of AI and real-world operations.
MSc European Politics (London School of Economics), Engineer's Diploma in Systems & Computers (Politehnica Bucharest). English C1-C2 (TOEFL 112/120).
In all honesty, Grenfell was caused by bad construction choices, materials and political penny-pinching. A second stairway filled with cyanide smoke would have had very limited benefits.
> OPV2 (specifically 2) cures zero infections because type 2 is extinct in the wild. So explain how my sentence isn't true?
OPV2 prevents cvdpv2 infections. This is a virus similar in all respects to wpv2, except its lower rate of paralysis onset. It's effectively the same disease, just like in my forest fire analogy.
Before the advent of nOPV2, OPV2 was much more effective at this than any other option. So you could not simply cease OPV2 production, post 2016 it was targeted to cvdpv2 hotspots only.
You have surely never operated a website that publishes user generated content. Every available textbox and file uploader will be filled with garbage so revolting that, without moderation, will quickly kill any chance of respectability for your site.
> presumably be exposed to an action for negligence if something went wrong as a result of them employing someone with that record.
That's just the type of bullshit that makes pizza restaurants not wanting to have a person with a criminal record anywhere in the building. It's a form of vigilante punishment that continues to for the life of a felon, way past the point where their debt to society has been supposedly paid.
Employers should be banned to ask or process such information. "Is currently wanted or on parole" - legitimate question, "was ever convicted" - No, you have no right to know that, except very limited cases defined by law: working with children and the vulnerable, large sums of cash, working in the financial sector etc.
I think 'ars' is leading you into a rhetorical trap. The objective was never eliminating the wild virus, the objective was curing polio. The OPV vaccine is a bit like fighting fire with fire: you burn down the forest (immunize susceptible hosts) in a controlled fashion, so that the forest fire cannot reach homes (paralyze children). But when your house burns down, it's irrelevant if it was the "wild" fire or a fire set by the firefighters, i.e a attenuated virus that mutated.
What 'ars' seems to be missing is that this particular "forest" is very rapidly growing back, in some countries you have in excess of 5% of the population as infants each year. Those are new hosts that were never vaccinated, and due to the extreme contagiousness of the disease a few years of lack of coverage can reignite the fire. What happened after OPV2 withdrawal was that a whole new generation inoculated with only bivalent (type 1+3) vaccine became susceptible to cvdpv2 that was still circulating in small pockets.
Essentially, the campaign failed the end-game strategy, they proved they can reduce the infection to arbitrarily low levels using trivalent OPV, but once you take OPV away, as they attempted for a single strain, the epidemic reignites. It's irrelevant if it's a wpv or cvdpv strain. The end-game was always considered a challenge by experts, but a variety of reasons, Covid, political issues etc. conspired to make it very difficult.
This whole thread leaves me very pessimistic about the prospects of eradication. If a relatively inteligent and educated internet-person that has proper sanitation cannot understand these epidemiological dynamics and claims that "OPV causes more infections than it cures", what's the chance you can explain it to rural farmers, especially after the global rise of the antivax movement after Covid?
> The reason is that the more you try to vaccinate the entire population, the more cases of cvdpv2 you cause. Until you hit that magic 100% you cause more harm than you prevent.
This makes little sense if you understand that both wpv2 and cvdpv2 were eliminated in the poorest regions of the world using OPV2. If what you claim is true, elimination would have been impossible, you would simply replace wpv2 with cvdpv2, since any attempt at eradication would seed new cvdpv2 cases.
Inactivated injectable vaccine, which does not boost herd immunity, is inefective in these countries with limited health systems.
> And in fact that people running this campaign noticed this and stopped vaccinating with opv2! (Which perhaps you did not know.)
Perhaps you refer to the global coordinated action to move from trivalent vaccine to bivalent (wpv1+ wpv3) in 2016 after the wild type 2 virus was certified as eradicated. But monovalent OPV2 was still being used recently to target specific areas where cvdpv2 is endemic. It makes little sense to use it elsewhere and seed cvdpv2.
> hot ass babe to pleasure me and raise nice children
Boy, are you in for a nasty surprise. But don't come crying to us, old men who warned you that what you really should want is a nice cabin near a lake with plenty of fish and an absolutely plain ass woman.
> if you give the vaccine, then the vaccine causes more cases (of type 2) than it's stopping.
> There's no solution here - either option makes things worse.
That's false and misleading. The OPV2 vaccine will stop both vaccine-derived cvdpv2 virus from spreading and the wild type 2, which has been eradicated. Since OPV2 is a live attenuated virus, it will also continue to multiply and confer protection to any person drinking contaminated water, the vaccine "spreads". It is thus highly effective at stopping polio and in no way it can be said that it "causes more cases than it's stopping".
The problem with OPV2 is that it has a relatively higher chance of reverting to an variant that causes polio. This is not a problem if the population has a high level of vaccination, since the live attenuated virus cannot propagate and mutate. It's only problematic in low coverage areas where it can multiply extensively in many hosts.
Thus, the cvdpv2 epidemic is an expression of low vaccination rates, similar in every way to a wild poliovirus resurgence. The novel OPV2 vaccine will improve the genetic stability of the attenuated virus, allowing further "viral vaccinations" in low coverage areas - but we could eradicate cvdpv2 today with the existing OPV2 vaccine, if only we could get good vaccinations rates everywhere, as it has happened in most of the world.
Nothing about the current tokamak based designs suggests they will be any cheaper to build for a given power rating than existing fission designs:
- They need large, highly advanced cryocooled superconducting magnets in very close proximity to a hundred million degree plasma. This only makes economic sense in massive, and expensive plants.
- They are a very strong source of fast neutrons useful to transmute cheap depleted uranium into plutonium, so carry massive proliferation risks, need close regulatory scrutiny and will require mounts of paperwork to operate, thus exceptionally inflexible to improvements and rapid iteration. Just like the current fission crop.
- Aneutronic fusion is a currently a purely theoretical concept, in the last 70 years nobody has been able to contain even the much cooler D-T plasma for economically viable durations and temperatures.
- The structure of the reactor becomes radiologically active and cleanup operations must be considered. Highly penetrating neutron radiation means some radiation will escape regardless of containment, requiring a radiological exclusion zone. No Mr. Fusion in your car, sorry.
- They operate and must breed sensitive nuclear materials - Tritium, a well known component of boosted thermonuclear weapons. The limited efficiency of tritium production from lithium-6 might require obtaining some from fission reactors to top up the fuel cycle and keep fusion reactors operating.
So when you draw the line, a life time of magnetic containment research has produced a speculative design that even if it were to work, which it doesn't, would be, in the best case scenario, comparable to existing fission designs that are being phased out for cost and risk issues.
A PhD money pit with zero chance of ever building anything useful.
> Proof of work mining uniquely only draws electricity when prices are economically viable, only buying electricity when there is excess supply
The numbers don't make sense to do that.
The average time a piece of hardware lives on the Bitcoin network is just 16 months, after which it becomes e-waste, too underpowered to matter.
So assuming the CAPEX is $1000 for a piece of hardware that draws 1kW, the most it can consume is 11 MWh during its lifetime - about $1500 at residential prices (Texas) or $500 at wholesale prices.
So just how deep can you throttle down the rig to take advantage of the low energy prices? The numbers seem to show that anything below 70-80% active ratio will start to cost you more in capital than you save in electricity.
So you can avoid the peaks but certainly cannot wait for the relatively rare excess renewables events.
Yup, a major breakthrough in this glacial field is doubling the energy output in 25 yeas.
So by the time captain Picard is born, we might have a very expensive and massive fusion reactor that will generate the same kind of energy we can generate today with very expensive and massive fission reactors.
Really now, this is pure garbage and does not solve the major problems in the field nor do most of the myriad startups trying to cash in on speculative seed funds.
Rooftop solar was always form of green subsidy: you get the same flat price for energy you dump into the network as the price the utility charges you. But what you put in at random times of your own choosing is much, much less valuable than a guaranteed power feed at any hour or season. At times it might have negative value, the power you put in costs the utility money. That simply cannot scale.
The only way I can see the two prices equal is if you provide power in the network on request from the utility, at specific time intervals from your own storage. But then you wouldn't need a power utility.
All totalitarians use social pressure to conform as a tool for suppressing dissidents. One day you are the only one with a green bubble in your group, and the next day, when your friends must name a likely spy for the greens, you know who they will think of.
So you either join the sinister cult willingly or you go to the gulag later, it's always like that, dammit.
"so the only ones willing to eat the risk are states."
Sure, if you can cook the accounting books to pay for a military nuclear industry, and also get a civil sector that has no need to pay for capital, insurance and cleanup. I doubt it's a scalable model, and France had its own share of fuckups and close calls.
Nuclear had roughly 60 years to deliver clean, cheap and reliable energy. If it did all that, no amount of green lobby could have derailed it. Unfortunately, and I feel that deeply as an engineer, it did not achieve those goals.
What we have currently have are slightly evolved designs over those of the 60s, which are fantastically expensive, take up to a decade to build and three more decades to recover investment, and present a large economic failure risk - see the Nukegate fiasco. No sane financial investor wants to approach nuclear, no one wants to insure it, so the only ones willing to eat the risk are states.
Yes, we had a window in the last 20 years to reinvent fission, have new, secure and cheap reactors which cannot proliferate, are passively safe and fast to deploy. The industry and regulators remained paralyzed, structurally risk averse. That window has now closed.
Renewables are cheaper then nuclear, fast to come online and have very low risk. Storage is quickly dropping in price. Together, storage and renewables, coupled with a smart grid, will be able to cover more and more of the demand and require fossils to be fired less and less. It does not matter if you go fosil for a whole week in a year where there is no sun, wind or rain, that's still only 2% of the production and you can use very expensive mitigations like CCS, as long as you have cheap renewables in the rest of the time.
Pro-tip: never connect any appliance, vehicle or device to the internet unless you really need the internet-enabled functionality. What you want is dumb devices that work reliably within their initial operating parameters and have a security air-gap.
This prevents exploitation of their crufty software, but also prevents the manufacturer to brick or otherwise use them against you with unscrupulous updates. If a software update exists, you will apply it manually if you are bothered by the specific bugs.
Never auto-update embedded software, never trust them on your network if they don't belong there, never trust their programmers to do anything but a half-ass hack, the absolute minimum required to get the product out the door, expecting they will issue updates.
Remote: Yes
Willing to relocate: Open to discuss
Technologies: Python, TypeScript, PHP, SQL, Rust, Bash. LLM integration RAG pipelines, agentic workflows. GCP (Cloud Run, Cloud Functions, Firestore,Cloud Storage), AWS (Lambda, S3, EC2), Azure. Docker, CI/CD, infrastructure-as-code. MySql, PostgreSQL, pgvector, Redis, Elasticsearch. Security hardening, GDPR. Hardware, low level control, electronics, embedded, robotics, networking and sysadmin etc.
Résumé/CV: https://www.linkedin.com/in/sbene/
Email: [email protected]
Two tracks in parallel across my career: building software and shaping policy. Started in embedded systems and DevOps (CI/CD, cloud-native infra, security auditing), crossed into government - advised the Mayor of Bucharest on a city-wide digital parking overhaul and served on the executive board of the municipal parking company. Now back to building: working part-time at an EdTech startup applying the modern AI stack (Python, TypeScript, LLM APIs, RAG, agentic workflows) to real products on GCP.
Engineering depth + public-sector leadership + hands-on AI. Looking for a role where technical ability and strategic thinking both matter - ideally at the intersection of AI and real-world operations.
MSc European Politics (London School of Economics), Engineer's Diploma in Systems & Computers (Politehnica Bucharest). English C1-C2 (TOEFL 112/120).