codesanitize

Raj Bakhru, Co-founder and CEO of BlueFlame AI – Interview Sequence

Robotics

codesanitize

-

2 April 2025

Raj Bakhru, Co-founder and CEO of BlueFlame AI – Interview Sequence

Raj Bakhru, Co-founder and CEO of BlueFlame AI, attracts on a wide-ranging background encompassing gross sales, advertising and marketing, software program improvement, company progress, and enterprise administration. All through his profession, he has performed a central function in growing top-tier instruments in different investments and cybersecurity.

Previously Chief Technique Officer at ACA, Raj oversaw company improvement and M&A, additionally serving as Interim Co-CEO, Chief Innovation Officer, and Head of RegTech and ESG. He was the founding father of Aponix, later ACA’s cyber division, a pacesetter within the alternate options sector. Raj’s expertise consists of roles as a quantitative software program developer at Kepos Capital, Highbridge, and Goldman Sachs Asset Administration. He holds a B.S. in Pc Engineering from Columbia College, together with CISSP and CFA credentials.

BlueFlame AI provides an AI-native, purpose-built, and LLM-agnostic answer designed for different funding managers.

The crew brings expertise throughout dealmaking, software program improvement, cybersecurity, and repair provision throughout the different funding sector. This background informs the corporate’s method to understanding industry-specific workflows and programs, permitting for the implementation of generative AI options tailor-made to the wants of different funding companies.

Are you able to share a bit about your background and the way your early experiences at Goldman Sachs, Highbridge, and Kepos Capital formed your understanding of know-how, cybersecurity, and different investments?

I spent an excellent a part of my early profession at quant funds, the place fashions traded the whole lot, from equities to FX to credit score and unique swaps. I realized an amazing quantity about how hedge funds work and the end-to-end workflows at hedge funds. Each have formed our later work in cybersecurity and now at BlueFlame tackling these workflows with AI. At ACA Group we realized the house’s compliance wants and constructed out the cyber packages for a whole lot of different funding advisers.

My background is consultant of our total crew: we now have 35+ of us with comparable however totally different experiences at a breadth of hedge funds, non-public fairness, and credit score outlets, and from distributors devoted to the house.

We imagine sensible, real-world expertise working on this house is crucial for translating AI proof-of-concepts into actuality for these companies.

What impressed you to transition from software program improvement in quantitative finance to entrepreneurship in cybersecurity and AI?

I’ve all the time been and nonetheless right now stay a technologist at coronary heart. The widespread thread throughout quantitative finance, cybersecurity, and AI is that on the time I used to be working within the house, it was present process a renaissance and large build-out. I deeply get pleasure from getting in on the bottom ground as a brand new house is evolving, serving to to show our shoppers and construct alongside them.

BlueFlame AI is designed particularly for different funding managers. What makes it totally different from common AI platforms like OpenAI’s ChatGPT or different enterprise AI options?

A vertical answer like BlueFlame isn’t actually a competitor to any horizontal options like ChatGPT. We offer an out-of-the-box set of options that make downside fixing sooner and simpler in our vertical, with extra particular tooling to deal with widespread use circumstances.

An instance is perhaps Funding Committee (IC) memo era. Whereas it is perhaps potential to immediate a horizontal answer to get a templated outcome, it received’t have the integrations to the CRMs, market information suppliers, or inner information to feed the IC memo. Horizontal options received’t have the power to drop the content material right into a template PowerPoint deck.

Are you able to stroll us by way of how BlueFlame AI enhances productiveness for hedge funds, non-public fairness companies, and different different buyers?

We implement AI-driven use circumstances for our shoppers, which regularly begin with front-office duties however can span your entire agency. These use circumstances, whereas widespread, range firm-to-firm. Some companies care lots about skilled community transcript summaries whereas others don’t do any. Some companies care lots about question credit score agreements whereas others don’t.

We work with our shoppers to establish the best ROI use case alternatives and deal with 3-5 of these of their first 12 months.

Given your in depth expertise in cybersecurity, what are the important thing safety dangers that different funding companies ought to concentrate on when adopting GenAI options?

Information safety and privateness are an enormous concern with GenAI utilization. First off, understanding the place your information goes and the way it’s being protected is paramount with LLM suppliers being hosted options. Subsequent, understanding the safeguards in place to guarantee that your information is safe and never getting used to coach fashions or inadvertently uncovered to different shoppers is crucial, as different funding companies take care of extremely delicate proprietary buying and selling methods and investor info that might be catastrophic if compromised. Lastly, companies should implement sturdy governance frameworks that embrace clear information dealing with insurance policies, common safety audits, and complete coaching packages to mitigate the chance and rising threats that would probably extract confidential info by way of interactions with these highly effective AI programs.

You’ve emphasised BlueFlame AI’s LLM-agnostic method. Why is that this an vital function, and the way does it profit your shoppers?

We imagine the facility of all of the LLMs collectively is larger than only one. We see that manifest day by day as we work with shoppers to construct out automations the place we all know one LLM may do higher than one other for a given activity. DeepSeek was an attention-grabbing second that confirmed open-source fashions have gotten extremely attention-grabbing and aggressive, too. Being LLM agnostic signifies that we will and can use all of them, our shoppers can accomplish that immediately with no need particular person licenses for every, and we will auto-route to the only option for a given activity on the given time. This continues to be helpful as fashions change over time.

Many companies wrestle with info overload. How does BlueFlame AI assist funding managers streamline analysis and due diligence?

BlueFlame helps with enterprise information administration by way of search and reply throughout all programs. We remedy for each info overload and information sprawl. A easy reply may reside in any one in every of a agency’s 5-10 programs. We glance throughout all of them to search out potential solutions to any given query inside their key programs and file shops.

Regulators are starting to pay shut consideration to AI utilization in monetary markets. How do you see compliance evolving within the AI-driven funding panorama?

In the present day, regulators count on insurance policies and procedures and considerate safety of investor information, particularly safety from 3^rd get together fashions coaching on delicate information). Shortly, we’ll see a compliance layer in opposition to brokers: these brokers might be “entry individuals” and must abide by the agency’s compliance guidelines like some other member of the crew.

What ought to hedge funds and personal fairness companies prioritize when integrating AI into their workflows whereas sustaining sturdy cybersecurity measures?

I feel when getting began, each agency ought to do two issues. First, establish the perfect use circumstances to your agency. Most frequently front-office duties ship the upper, extra speedy ROI. Map these use circumstances in opposition to capabilities out there available in the market to establish the 3-5 you wish to lean in on. Second, establish the appropriate product and companion.Discover a agency you suppose might be responsive and capable of iterate with you—one with confirmed success and the appropriate cyber/privateness/compliance posture.

What does the way forward for AI in different investments appear like? Do you see AI ultimately enjoying a task in making funding choices?

AI is already concerned in funding decision-making, however that is solely changing into extra commonplace. Many PE features can have AI brokers, like a sourcing agent to assist with goal outreach and scheduling. Ultimately, there might be quantitative PE companies that function totally with AI fashions as quantitative hedge funds do. These quant PE companies can have AI brokers interacting with bankers, legal professionals, and so on. to finish offers.

Thanks for the good interview, readers who want to study extra ought to go to BlueFlame AI.

Tesla Board Asks Elon Musk to Step Down

Green Technology

codesanitize

-

2 April 2025

0

Join day by day information updates from CleanTechnica on e mail. Or comply with us on Google Information!

It’s broadly repeated that Winston Churchill as soon as stated, “You’ll be able to at all times depend on People to do the fitting factor — after they’ve tried all the things else.” Though it’s not clear if Churchill ever did really say that, it has grow to be a well-liked quote and has offered numerous laughs. Nevertheless it’s not solely People who battle to do the fitting factor the primary time, and it’s principally a part of human nature that it may possibly take us a very long time to just accept arduous truths and take care of issues and challenges realistically — much more so when one thing important has really modified and gone from useful to dangerous.

It’s no secret that Tesla gross sales have been half a world off of the place they have been alleged to be. Just a few years in the past, Tesla CEO Elon Musk stated that Tesla gross sales would develop at about 50% a 12 months by means of the 2020s — not essentially yearly, however on common. That was taking place for a bit, however issues haven’t precisely been going in keeping with plan for greater than a 12 months now. Removed from seeing 50% gross sales development in 2024, Tesla gross sales really declined final 12 months. That is within the context of the broader EV market rising. Development was maybe not as excessive as many anticipated, however EV gross sales have been rising — simply not Tesla’s.

Prior to now 12 months, Musk has gone from pretty politically impartial and indifferent to extraordinarily politically concerned — and sometimes to a surprising diploma. His extremist tendencies have dramatically upset tens of millions of individuals within the US, the UK, Germany, France, and lots of, many different locations. This has all led to important gross sales hits, in addition to protests at Tesla shops, Tesla house owners placing bumper stickers on their automobiles to disassociate themselves from Elon Musk, and even unlawful vandalism and violence that the majority of us would by no means assist or condone. Whenever you’re attending to the purpose of protests and bumper stickers, you realize that there’s additionally obtained to be some extent of market response. Actually, Tesla gross sales are down way more in 2025 than they even have been in 2024.

After all, the refreshed Tesla Mannequin Y shook issues up a bit — manufacturing strains needed to be shut down and reworked, some consumers have been ready on the brand new Mannequin Y, and so on., and so on. Nevertheless it appears extremely unlikely that has prompted as a lot of a dip in gross sales as we’ve seen, particularly given every kind of different proof that Musk’s life on the White Home, in DOGE workplaces, and sometimes proper beside or behind Donald Trump is hurting Tesla gross sales.

Then there’s the virtually fixed tweeting (displaying clear indicators of sleep deprivation), preoccupation with different corporations (Elon Musk’s X merging with Elon Musk’s xAI … SpaceX rockets exploding whereas others carry astronauts house), drama with the mama of his eleventh (or twelfth?) child, quite a few lawsuits, and the distracts simply go on and on and on. Any affordable and impartial board of administrators of a Fortune 500 firm would have pulled the CEO into the board room a very long time in the past and stated, “hey, you might want to get targeted on our firm and bettering these numbers otherwise you’ve obtained to go.” (By the best way, we simply skipped previous virtually a decade of missed self-driving targets.)

Effectively, the Tesla board of administrators isn’t precisely affordable and impartial. Musk has been allowed to tarnish the Tesla model (that he as soon as was vital in increase) and drive gross sales down, down, down because the board has remained silent and seemingly ineffective. Till at this time….

Information has damaged that the Tesla board of administrators has requested Elon Musk to step down from his position as CEO. He would nonetheless have a job within the firm, however not operating it and never being the principle face of the corporate. It’s not clear but who will step in as the brand new CEO, although. Keep tuned — we should always have extra information quickly. That stated, when the clock strikes midnight, the fairytale may very well be over, and we may very well be again the place we began — exhausting all different choices earlier than going with the plain one.

Whether or not you have got solar energy or not, please full our newest solar energy survey.

Chip in a number of {dollars} a month to assist assist impartial cleantech protection that helps to speed up the cleantech revolution!

Have a tip for CleanTechnica? Need to promote? Need to counsel a visitor for our CleanTech Discuss podcast? Contact us right here.

Join our day by day publication for 15 new cleantech tales a day. Or join our weekly one if day by day is simply too frequent.

Commercial

CleanTechnica makes use of affiliate hyperlinks. See our coverage right here.

CleanTechnica’s Remark Coverage

ios – separate presentation logic from the view in SwiftUI

iOS Development

codesanitize

-

2 April 2025

0

I have been making an attempt to separate presentation logic from the view itself and I discovered this sample that initially look appeared to work.

struct MyView: View {
   @StateObject var viewModel = ViewModel()

   ...
}

extension MyView {
   class ViewModel: ObservableObject {
      ...
   }
}

This works properly besides when the view is determined by a dependency owned by the mother or father view. StateObject documentation provides the next instance:

struct MyInitializableView: View {
    @StateObject non-public var mannequin: DataModel


    init(title: String) {
        // SwiftUI ensures that the next initialization makes use of the
        // closure solely as soon as in the course of the lifetime of the view, so
        // later adjustments to the view's title enter don't have any impact.
        _model = StateObject(wrappedValue: DataModel(title: title))
    }


    var physique: some View {
        VStack {
            Textual content("Title: (mannequin.title)")
        }
    }
}

Nevertheless, they instantly warn that this strategy solely works if the exterior knowledge would not change. In any other case the information mannequin will not have entry to up to date values in any of the properties.

Within the above instance, if the title enter to MyInitializableView
adjustments, SwiftUI reruns the view’s initializer with the brand new worth.
Nevertheless, SwiftUI runs the autoclosure that you simply present to the state
object’s initializer solely the primary time you name the state object’s
initializer, so the mannequin’s saved title worth doesn’t change. What
could be one of the simplest ways to separate presentation logic from the view
itself? Subscribing to publishers in a use case, calculating body
sizes, logic to find out whether or not a toddler view is seen or not, and so forth
could be higher off in a unique file that the view makes use of to attract
itself.

To keep away from having an excessive amount of logic within the view like this:

NOTE: This has nice efficiency advantages since any updates to particular person will trigger a re-render WITHOUT inflicting your entire view to be reinitialised. Its lifecycle isn’t affected

struct PersonView: View {
    let particular person: Particular person
    
    non-public let dateFormatter = DateFormatter()
    
    var physique: some View {
        VStack(alignment: .main) {
            Textual content(fullName)
            Textual content(birthday)
        }
    }
    
    var fullName: String {
        "(particular person.firstName) (particular person.lastName)"
    }
    
    var birthday: String {
        dateFormatter.dateFormat = "MMM d"
        
        return dateFormatter.string(from: particular person.dateOfBirth)
    }
}

We may separate the presentation logic for the view’s rendering like this:

struct PersonView: View {
    @StateObject non-public var viewModel: ViewModel
    
    init(particular person: Particular person) {
        self._viewModel = .init(wrappedValue: ViewModel(particular person: particular person))
    }
    
    var physique: some View {
        VStack(alignment: .main) {
            Textual content(viewModel.fullName)
            Textual content(viewModel.birthday)
        }
    }
}

extension PersonView {
    class ViewModel: ObservableObject {
        let particular person: Particular person
        
        non-public let dateFormatter: DateFormatter
        
        init(particular person: Particular person) {
            self.particular person = particular person
            
            let dateFormatter = DateFormatter()
            dateFormatter.dateFormat = "MMM d"
            
            self.dateFormatter = dateFormatter
        }
        
        var fullName: String {
            "(particular person.firstName) (particular person.lastName)"
        }
        
        var birthday: String {
            dateFormatter.string(from: particular person.dateOfBirth)
        }
    }
}

Nevertheless, as talked about within the documentation any updates to any of Particular person’s properties will not be mirrored within the view.

There are a number of methods to drive reinitialisation by altering the view’s id, however all of them include efficiency points and different unwanted effects.

Be aware of the efficiency price of reinitializing the state object
each time the enter adjustments. Additionally, altering view id can have
unwanted effects. For instance, SwiftUI doesn’t routinely animate
adjustments contained in the view if the view’s id adjustments on the similar
time. Additionally, altering the id resets all state held by the view,
together with values that you simply handle as State, FocusState, GestureState,
and so forth.

Is there a approach to obtain a extra clear separation of issues whereas nonetheless leveraging SwftUI’s optimisations when re-rendering views?

New modelling instrument may assist defend wildlife from offshore developments

Green Technology

codesanitize

-

2 April 2025

0

New modelling instrument may assist defend wildlife from offshore developments

A brand new useful resource has been developed in a bid to raised defend seabirds from the impacts of offshore wind farms. The group behind it says the modern modelling instrument has the potential to save lots of wildlife, whereas making certain the efficient and sustainable improvement of renewable vitality sources.

Led by researchers on the College of Glasgow, the brand new modelling instrument, printed within the journal Strategies in Ecology and Evolution, is the primary of its variety to precisely predict house use of seabird colonies with out requiring in depth satellite tv for pc monitoring information, which is commonly not obtainable.

Seabird environmental evaluation instruments are used to tell planning choices for offshore windfarms. Whereas helpful, present evaluation strategies can differ in accuracy and should result in offshore windfarms being in-built areas with excessive fowl density. Conversely, offshore windfarm developments could also be inadvertently rejected primarily based on overestimates of seabirds at sea.

Many seabird species nest in colonies on small items of land, together with clusters of rocks off the coast. From there, birds fly and forage round a neighborhood space at sea – their house vary – which varies in dimension relying on colony dimension and site. Due to these attribute behaviours, central-place foragers similar to seabirds, are significantly delicate to environmental stressors of their native environment, together with the event of wind farms, which can have extreme impacts on colony numbers and wellbeing.

Utilizing GPS monitoring information from 8 Northern gannet colonies to confirm their predictions, the researchers present that their new instrument roughly doubles on the predictive energy of different business normal strategies. The brand new instrument was 73% correct on common, compared to 41% and 31% accuracy of present seabird evaluation instruments. Present strategies additionally over- and underestimated colony publicity to offshore wind farms in several eventualities.

Researchers imagine their new instrument could possibly be transformative for offshore windfarm planning, permitting us to guard wildlife whereas additionally safely constructing sources of renewable vitality.

Lead writer of the examine, PhD scholar Holly Niven from the Faculty of Biodiversity, One Well being & Veterinary Drugs, stated: “Correct estimation of the impacts of offshore wind farms and different stressors on seabirds may help us make extra knowledgeable choices about offshore wind farm plans and defend the species dwelling round our coasts.”

Moreover, researchers say the brand new modelling instrument may be used to precisely predict the house use of different colonial wildlife, together with seals, bats and bees.

Jason Matthiopoulos, Professor of Spatial and Inhabitants Ecology who supervised the examine, stated: “Paradoxically, totally different environmentally constructive actions similar to wildlife conservation and our progress in the direction of inexperienced vitality can come into battle with one another. Resolving these conflicts depends on good information, however equally, on state-of-the-art pc modelling strategies.”

Jana Jeglinski, Analysis Fellow and co-supervisor of the examine, stated: “Many seabird colonies are positioned at distant islands or cliffs that make GPS monitoring research extraordinarily difficult and even not possible. Our methodology can predict biologically life like house ranges and publicity for such inaccessible colonies and it will probably additionally forecast future house ranges given the dimensions of a colony – that is vital since offshore windfarm development will drastically improve within the close to future.”

The examine, ‘In the direction of biologically life like estimates of house vary and spatial publicity for colonial animals’ is printed in Strategies in Ecology and Evolution. The work was funded by the UK Authorities Division for Vitality Safety & Internet Zeros Offshore Vitality Strategic Environmental Evaluation OESEA program.

Lucid PhaaS Hits 169 Targets in 88 International locations Utilizing iMessage and RCS Smishing

Hacking

codesanitize

-

2 April 2025

0

Lucid PhaaS Hits 169 Targets in 88 International locations Utilizing iMessage and RCS Smishing

A brand new refined phishing-as-a-service (PhaaS) platform referred to as Lucid has focused 169 entities in 88 nations utilizing smishing messages propagated through Apple iMessage and Wealthy Communication Providers (RCS) for Android.

Lucid’s distinctive promoting level lies in its weaponizing of legit communication platforms to sidestep conventional SMS-based detection mechanisms.

“Its scalable, subscription-based mannequin permits cybercriminals to conduct large-scale phishing campaigns to reap bank card particulars for monetary fraud,” Swiss cybersecurity firm PRODAFT mentioned in a technical report shared with The Hacker Information.

“Lucid leverages Apple iMessage and Android’s RCS expertise, bypassing conventional SMS spam filters and considerably rising supply and success charges.”

Lucid is assessed to be the work of a Chinese language-speaking hacking crew referred to as the XinXin group (aka Black Expertise), with the phishing campaigns primarily focusing on Europe, the UK, and the USA with an intent to steal bank card knowledge and personally identifiable data (PII).

The menace actors behind the service, extra importantly, have developed different PhaaS platforms like Lighthouse and Darcula, the latter of which has been up to date with capabilities to clone any model’s web site to create a phishing model. The developer of Lucid is a menace actor codenamed LARVA-242, who can be a key determine within the XinXin group.

All three PhaaS platforms share overlaps in templates, goal swimming pools, and techniques, alluding to a flourishing underground financial system the place Chinese language-speaking actors are leveraging Telegram to promote their warez on a subscription foundation for profit-driven motives.

Phishing campaigns counting on these companies have been discovered to impersonate postal companies, courier corporations, toll cost methods, and tax refund businesses, using convincing phishing templates to deceive victims into offering delicate data.

The big-scale actions are powered on the backend through iPhone machine farms and cellular machine emulators working on Home windows methods to ship tons of of hundreds of rip-off messages containing bogus hyperlinks in a coordinated trend. The cellphone numbers to be focused are acquired by varied strategies reminiscent of knowledge breaches and cybercrime boards.

“For iMessage’s link-clicking restrictions, they make use of ‘please reply with Y’ strategies to determine two-way communication,” PRODAFT defined. “For Google’s RCS filtering, they continually rotate sending domains/numbers to keep away from sample recognition.”

“For iMessage, this entails creating momentary Apple IDs with impersonated show names, whereas RCS exploitation leverages service implementation inconsistencies in sender verification.”

In addition to providing automation instruments that simplify the creation of customizable phishing web sites, the pages themselves incorporate superior anti-detection and evasion strategies like IP blocking, user-agent filtering, and time-limited single-use URLs.

Lucid additionally helps the power to observe sufferer exercise and document each single interplay with the phishing hyperlinks in real-time through a panel, permitting its clients to extract the entered data. Bank card particulars submitted by victims are subjected to extra verification steps. The panel is constructed utilizing the open-source Webman PHP framework.

“The Lucid PhaaS panel has revealed a extremely organized and interconnected ecosystem of phishing-as-a-service platforms operated by Chinese language-speaking menace actors, primarily underneath the XinXin group,” the corporate mentioned.

“The XinXin group develops and makes use of these instruments and income from promoting stolen bank card data whereas actively monitoring and supporting the event of comparable PhaaS companies.”

It is price noting that the findings from PRODAFT mirror that of Palo Alto Networks Unit 42, which not too long ago referred to as out unspecified menace actors for using the area sample “com-” to register over 10,000 domains for propagating varied SMS phishing scams through Apple iMessage.

The event comes as Barracuda warned of a “huge spike” in PhaaS assaults in early 2025 utilizing Tycoon 2FA, EvilProxy, and Sneaky 2FA, with every service accounting for 89%, 8%, and three% of all of the PhaaS incidents, respectively.

“Phishing emails are the gateway for a lot of assaults, from credential theft to monetary fraud, ransomware, and extra,” Barracuda safety researcher Deerendra Prasad mentioned. “The platforms that energy phishing-as-a-service are more and more complicated and evasive, making phishing assaults each more durable for conventional safety instruments to detect and extra highly effective when it comes to the harm they’ll do.”

Discovered this text fascinating? Comply with us on Twitter and LinkedIn to learn extra unique content material we submit.

Raj Bakhru, Co-founder and CEO of BlueFlame AI – Interview Sequence

Tesla Board Asks Elon Musk to Step Down

ios – separate presentation logic from the view in SwiftUI

New modelling instrument may assist defend wildlife from offshore developments

Lucid PhaaS Hits 169 Targets in 88 International locations Utilizing iMessage and RCS Smishing

ABOUT US