Social
Automated reasoning

A gentle introduction to automated reasoning

Meet Amazon Science’s newest research area.

By Byron Cook
December 1, 2021
18 min read
Share

This week, Amazon Science added automated reasoning to its list of research areas. We made this change because of the impact that automated reasoning is having here at Amazon. For example, Amazon Web Services’ customers now have direct access to automated-reasoning-based features such as IAM Access Analyzer, S3 Block Public Access, or VPC Reachability Analyzer. We also see Amazon development teams integrating automated-reasoning tools into their development processes, raising the bar on the security, durability, availability, and quality of our products.

The goal of this article is to provide a gentle introduction to automated reasoning for the industry professional who knows nothing about the area but is curious to learn more. All you will need to make sense of this article is to be able to read a few small C and Python code fragments. I will refer to a few specialist concepts along the way, but only with the goal of introducing them in an informal manner. I close with links to some of our favorite publicly available tools, videos, books, and articles for those looking to go more in-depth.

Let’s start with a simple example. Consider the following C function:

bool f(unsigned int x, unsigned int y) {
   return (x+y == y+x);
}



        
Take a few moments to answer the question “Could f ever return false?” This is not a trick question: I’ve purposefully used a simple example to make a point.
To check the answer with exhaustive testing, we could try executing the following doubly nested test loop, which calls f on all possible pairs of values of the type unsigned int:

            

    
#include<stdio.h>
#include<stdbool.h>
#include<limits.h>

bool f(unsigned int x, unsigned int y) {
   return (x+y == y+x);
}

void main() {
   for (unsigned int x=0;1;x++) {
      for (unsigned int y=0;1;y++) {
         if (!f(x,y)) printf("Error!\n");
         if (y==UINT_MAX) break;
      }
      if (x==UINT_MAX) break;
   }
}



        
Unfortunately, even on modern hardware, this doubly nested loop will run for a very long time. I compiled it and ran it on a 2.6 GHz Intel processor for over 48 hours before giving up.
Why does testing take so long? Because UINT_MAX is typically 4,294,967,295, there are 18,446,744,065,119,617,025 separate f calls to consider. On my 2.6 GHz machine, the compiled test loop called f approximately 430 million times a second. But to test all 18 quintillion cases at this performance, we would need over 1,360 years.
When we show the above code to industry professionals, they almost immediately work out that f can't return false as long as the underlying compiler/interpreter and hardware are correct. How do they do that? They reason about it. They remember from their school days that x + y can be rewritten as y + x and conclude that f always returns true.

            

    
        






    
    

        
            
Re:Invent 2021 keynote address by Peter DeSantis, senior vice president for utility computing at Amazon Web Services

        
        
            
Skip to 15:49 for a discussion of Amazon Web Services' work on automated reasoning.

        
    



        
An automated reasoning tool does this work for us: it attempts to answer questions about a program (or a logic formula) by using known techniques from mathematics. In this case, the tool would use algebra to deduce that x + y == y + x can be replaced with the simple expression true.
Automated-reasoning tools can be incredibly fast, even when the domains are infinite (e.g., unbounded mathematical integers rather than finite C ints). Unfortunately, the tools may answer “Don’t know” in some instances. We'll see a famous example of that below.
The science of automated reasoning is essentially focused on driving the frequency of these “Don’t know” answers down as far as possible: the less often the tools report "Don't know" (or time out while trying), the more useful they are.
Today’s tools are able to give answers for programs and queries where yesterday’s tools could not. Tomorrow’s tools will be even more powerful. We are seeing rapid progress in this field, which is why at Amazon, we are increasingly getting so much value from it. In fact, we see automated reasoning forming its own Amazon-style virtuous cycle, where more input problems to our tools drive improvements to the tools, which encourages more use of the tools.
A slightly more complex example. Now that we know the rough outlines of what automated reasoning is, the next small example gives a slightly more realistic taste of the sort of complexity that the tools are managing for us.

            

    
void g(int x, int y) {
   if (y > 0)
      while (x > y)
         x = x - y;
}



        
Or, alternatively, consider a similar Python program over unbounded integers:

            

    
def g(x, y):
   assert isinstance(x, int) and isinstance(y, int)
   if y > 0:
      while x > y:
         x = x - y



        
Try to answer this question: “Does g always eventually return control back to its caller?”
When we show this program to industry professionals, they usually figure out the right answer quickly. A few, especially those who are aware of results in theoretical computer science, sometimes mistakenly think that we can't answer this question, with the rationale “This is an example of the halting problem, which has been proved insoluble”. In fact, we can reason about the halting behavior for specific programs, including this one. We’ll talk more about that later.
Here’s the reasoning that most industry professionals use when looking at this problem:
  1. In the case where y is not positive, execution jumps to the end of the function g. That’s the easy case.
  2. If, in every iteration of the loop, the value of the variable x decreases, then eventually, the loop condition x > y will fail, and the end of g will be reached.
  3. The value of x always decreases only if y is always positive, because only then does the update to x (i.e., x = x - y) decrease x. But y’s positivity is established by the conditional expression, so x always decreases.
The experienced programmer will usually worry about underflow in the x = x - y command of the C program but will then notice that x > y before the update to x and thus cannot underflow.
If you carried out the three steps above yourself, you now have a very intuitive view of the type of thinking an automated-reasoning tool is performing on our behalf when reasoning about a computer program. There are many nitty-gritty details that the tools have to face (e.g., heaps, stacks, strings, pointer arithmetic, recursion, concurrency, callbacks, etc.), but there’s also decades of research papers on techniques for handling these and other topics, along with various practical tools that put these ideas to work.

            

    

        
    
        Policy-code.gif
    


        

        
    


    
        
Automated reasoning can be applied to both policies (top) and code (bottom). In both cases, an essential step is reasoning about what's always true.

    


        
The main takeaway is that automated-reasoning tools are usually working through the three steps above on our behalf: Item 1 is reasoning about the program’s control structure. Item 2 is reasoning about what is eventually true within the program. Item 3 is reasoning about what is always true in the program.
Note that configuration artifacts such as AWS resource policies, VPC network descriptions, or even makefiles can be thought of as code. This viewpoint allows us to use the same techniques we use to reason about C or Python code to answer questions about the interpretation of configurations. It’s this insight that gives us tools like IAM Access Analyzer or VPC Reachability Analyzer.
An end to testing?
As we saw above when looking at f and g, automated reasoning can be dramatically faster than exhaustive testing. With tools available today, we can show properties of f or g in milliseconds, rather than waiting lifetimes with exhaustive testing.
Can we throw away our testing tools now and just move to automated reasoning? Not quite. Yes, we can dramatically reduce our dependency on testing, but we will not be completely eliminating it any time soon, if ever. Consider our first example:

            

    
bool f(unsigned int x, unsigned int y) {
   return (x + y == y + x);
}



        
Recall the worry that a buggy compiler or microprocessor could in fact cause an executable program constructed from this source code to return false. We might also need to worry about the language runtime. For example, the C math library or the Python garbage collector might have bugs that cause a program to misbehave.
What’s interesting about testing, and something we often forget, is that it’s doing much more than just telling us about the C or Python source code. It’s also testing the compiler, the runtime, the interpreter, the microprocessor, etc. A test failure could be rooted in any of those tools in the stack.
Automated reasoning, in contrast, is usually applied to just one layer of that stack — the source code itself, or sometimes the compiler or the microprocessor. What we find so valuable about reasoning is it allows us to clearly define both what we do know and what we do not know about the layer under inspection.
Furthermore, the models of the surrounding environment (e.g., the compiler or the procedure calling our procedure) used by the automated-reasoning tool make our assumptions very precise. Separating the layers of the computational stack helps make better use of our time, energy, and money and the capabilities of the tools today and tomorrow.
Unfortunately, we will almost always need to make assumptions about something when using automated reasoning — for example, the principles of physics that govern our silicon chips. Thus, testing will never be fully replaced. We will want to perform end-to-end testing to try and validate our assumptions as best we can.
An impossible program
I previously mentioned that automated-reasoning tools sometimes return “Don’t know” rather than “yes” or “no”. They also sometimes run forever (or time out), thus never returning an answer. Let’s look at the famous "halting problem" program, in which we know tools cannot return “yes” or “no”.
Imagine that we have an automated-reasoning API, called terminates, that returns “yes” if a C function always terminates or “no” when the function could execute forever. As an example, we could build such an API using the tool described here (shameless self-promotion of author’s previous work). To get the idea of what a termination tool can do for us, consider two basic C functions, g (from above),

            

    
void g(int x, int y) {
   if (y > 0)
      while (x > y)
         x = x - y;
}



        
and g2:

            

    
void g2(int x, int y) {
   while (x > y)
      x = x - y;
}



        
For the reasons we have already discussed, the function g always returns control back to its caller, so terminates(g) should return true. Meanwhile, terminates(g2) should return false because, for example, g2(5, 0) will never terminate.
Now comes the difficult function. Consider h:

            

    
void h() {
   if terminates(h) while(1){}
}



        
Notice that it's recursive. What’s the right answer for terminates(h)? The answer cannot be "yes". It also cannot be "no". Why?
Imagine that terminates(h) were to return "yes". If you read the code of h, you’ll see that in this case, the function does not terminate because of the conditional statement in the code of h that will execute the infinite loop while(1){}. Thus, in this case, the terminates(h) answer would be wrong, because h is defined recursively, calling terminates on itself.
Similarly, if terminates(h) were to return "no", then h would in fact terminate and return control to its caller, because the if case of the conditional statement is not met, and there is no else branch. Again, the answer would be wrong. This is why the “Don’t know” answer is actually unavoidable in this case.
The program h is a variation of examples given in Turing’s famous 1936 paper on decidability and Gödel’s incompleteness theorems from 1931. These papers tell us that problems like the halting problem cannot be “solved”, if by“solved” we mean that the solution procedure itself always terminates and answers either “yes” or “no” but never “Don’t know”. But that is not the definition of “solved” that many of us have in mind. For many of us, a tool that sometimes times out or occasionally returns “Don’t know” but, when it gives an answer, always gives the right answer is good enough.
This problem is analogous to airline travel: we know it’s not 100% safe, because crashes have happened in the past, and we are sure that they will happen in the future. But when you land safely, you know it worked that time. The goal of the airline industry is to reduce failure as much as possible, even though it’s in principle unavoidable.
To put that in the context of automated reasoning: for some programs, like h, we can never improve the tool enough to replace the "Don't know" answer. But there are many other cases where today's tools answer "Don't know", but future tools may be able to answer "yes" or "no". The modern scientific challenge for automated-reasoning subject-matter experts is to get the practical tools to return “yes” or “no” as often as possible. As an example of current work, check out CMU professor and Amazon Scholar Marijn Heule and his quest to solve the Collatz termination problem.
Another thing to keep in mind is that automated-reasoning tools are regularly trying to solve “intractable” problems, e.g., problems in the NP complexity class. Here, the same thinking applies that we saw in the case of the halting problem: automated-reasoning tools have powerful heuristics that often work around the intractability problem for specific cases, but those heuristics can (and sometimes do) fail, resulting in “Don’t know” answers or impractically long execution time. The science is to improve the heuristics to minimize that problem.
Nomenclature
A host of names are used in the scientific literature to describe interrelated topics, of which automated reasoning is just one. Here’s a quick glossary:
  • A logic is a formal and mechanical system for defining what is true and untrue. Examples: propositional logic or first-order logic.
  • A theorem is a true statement in logic. Example: the four-color theorem.
  • A proof is a valid argument in logic of a theorem. Example: Gonthier's proof of the four-color theorem. 
  • A mechanical theorem prover is a semi-automated-reasoning tool that checks a machine-readable expression of a proof often written down by a human. These tools often require human guidance. Example: HOL-light, from Amazon researcher John Harrison. 
  • Formal verification is the use of theorem proving when applied to models of computer systems to prove desired properties of the systems. Example: the CompCert verified C compiler. 
  • Formal methods is the broadest term, meaning simply the use of logic to reason formally about models of systems. 
  • Automated reasoning focuses on the automation of formal methods. 
  • A semi-automated-reasoning tool is one that requires hints from the user but still finds valid proofs in logic. 
As you can see, we have a choice of monikers when working in this space. At Amazon, we’ve chosen to use automated reasoning, as we think it best captures our ambition for automation and scale. In practice, some of our internal teams use both automated and semi-automated reasoning tools, because the scientists we've hired can often get semi-automated reasoning tools to succeed where the heuristics in fully automated reasoning might fail. For our externally facing customer features, we currently use only fully automated approaches.
Next steps
In this essay, I’ve introduced the idea of automated reasoning, with the smallest of toy programs. I haven’t described how to handle realistic programs, with heap or concurrency. In fact, there are a wide variety of automated-reasoning tools and techniques, solving problems in all kinds of different domains, some of them quite narrow. To describe them all and the many branches and sub-disciplines of the field (e.g. “SMT solving”, “higher-order logic theorem proving”, “separation logic”) would take thousands of blogs posts and books.
Automated reasoning goes back to the early inventors of computers. And logic itself (which automated reasoning attempts to solve) is thousands of years old. In order to keep this post brief, I’ll stop here and suggest further reading. Note that it’s very easy to get lost in the weeds reading depth-first into this area, and you could emerge more confused than when you started. I encourage you to use a bounded depth-first search approach, looking sequentially at a wide variety of tools and techniques in only some detail and then moving on, rather than learning only one aspect deeply.
Suggested books:
International conferences/workshops:
Tool competitions:
Some tools:
Interviews of Amazon staff about their use of automated reasoning:
AWS Lectures aimed at customers and industry:
AWS talks aimed at the automated-reasoning science community:
AWS blog posts and informational videos:
Some course notes by Amazon Scholars who are also university professors:
A fun deep track:
Some algorithms found in the automated theorem provers we use today date as far back as 1959, when Hao Wang used automated reasoning to prove the theorems from Principia Mathematica.

    



                            
                        

                    
                


                
                    

                        

    

    

        Research areas
    


    
        
    



                    

                

                
                    

                        

    

    

        Tags
    


    
        
    



                    

                

                

                
                    
About the Author

                    

                        
                            
    

        

            
                

                    Byron Cook
                

            
            
            
                

                    Amazon vice president and distinguished scientist Byron Cook is a leader in the field of formal verification, known for his contributions to SAT, SMT, and symbolic model checking, with applications to biological systems, computer operating systems, programming languages, and security. Byron’s work on automated reasoning at Amazon has led to higher levels of assurance in the cloud as well as new customer features.
                

            
        

    



                        
                    

                
            

        


        
    

    
    



    

        
Related content

        
    




    
        
    

    



    
    



    

        
Work with us
See more jobs
        
            See more jobs
        
    




    
        

            

                
                    

                        

    

    

        
            

                Applied Scientist I, Amazon Shipping
            

        

        

            
                
                    

                
            

            

            

            

            
                
IN, HR, Gurugram

            
        


        
    
Building large-scale forecasting and optimization systems that power Amazon’s global transportation network and directly impact customer experience and cost. Key job responsibilities 1. Guide model and system design across a range of techniques, including tree-based models, deep learning (LSTMs, transformers), LLMs, and reinforcement learning. 2. Ensure models are production-ready, scalable, and robust through close partnership with stakeholders. 3. Partner with Product, Operations, and Engineering leaders to enable proactive decision-making and corrective actions. 4 Own end-to-end business metrics, directly influencing customer experience, cost optimization, and network reliability. 5. Help contribute to the broader ML community through publications, conference submissions, and internal knowledge sharing.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Economist, Pricing Science
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, WA, Seattle

            
        


        
    
Estimating the demand response of a pricing decision is genuinely hard. The causal effects are delayed, noisy, and confounded by factors that standard experiment analysis wasn't designed to handle. Most pricing teams default to heuristics not because they don't care about customer responses, but because measuring them rigorously is an unsolved problem. P2OS is building the science to solve it. We're hiring an Economist to own that work — defining how we estimate digital demand response in a pricing context, building the identification strategies that make those estimates credible, and translating outputs into something pricing teams can use to make better decisions. The role sits at the intersection of econometric methodology and production-quality analysis, and requires someone who can operate independently in both. As science lead, you'll own the digital pricing methodology domain, and be the internal authority on causal inference for pricing across P2OS and partner teams. Key job responsibilities * Own the end-to-end digital pricing methodology for pricing — identification strategy, modeling choices, validation approach, and business use cases — and drive adoption across pricing contexts * Deliver high-stakes analyses connecting digital pricing estimates to a concrete pricing decision and strategy change at VP+ level * Apply advanced causal methods to live pricing problems; document approaches so the team can build on and extend them. * Provide causal inference guidance on pricing experiment questions as they arise — being the methodology resource when experiments generate relevant questions * Serve as cross-team economic advisor to Digital Finance, Customer Behavior, and Demand Science on assumptions and causal identification * Actively mentor junior scientists, earn trust of cross-functional tech and product partners. A day in the life In a typical day, you'll move between methodology work and stakeholder-facing analysis. - On the science side, that means reviewing identification assumptions with the Causal AS, validating estimation choices for the LTV framework, and documenting methodology decisions in ways that non-economists can act on. - On the applied side, you'll be in rooms with Finance, Pricing PMs, and other science teams: aligning on LTV definitions, resolving disagreements between competing metrics, and translating causal findings into recommendations that land in strategy reviews. - As tech lead, you need to work to develop the economists and scientists on your scrum: structured reviews, identification strategy feedback, and raising the quality of analyses before they reach stakeholders. The mix shifts, but the through-line is to progress the LTV methodology from open questions to shipped frameworks, and making sure the team's causal work is rigorous enough to hold up when it counts. About the team P2Optimization Science (P2OS) is responsible for the ML models and analytical frameworks that drive pricing decisions at scale. The team spans demand lift modeling, pricing error detection, customer lifetime value, and experimentation. Our small team of specialized applied scientists and economists works closely alongside engineers, and pricing product managers.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Applied Scientist II, Amazon, Amazon
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, WA, Seattle

            
        


        
    
We’re working to improve shopping on Amazon using the conversational capabilities of large language models, and are searching for pioneers who are passionate about technology, innovation, and customer experience, and are ready to make a lasting impact on the industry. You'll be working with talented scientists, engineers, and technical program managers (TPM) to innovate on behalf of our customers. If you're fired up about being part of a dynamic, driven team, then this is your moment to join us on this exciting journey!




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Sr. Applied Scientist,  Amazon Robotics
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, MA, Boston

            
        


        
    
Are you interested in how to build AI reasoning systems that give provably correct answers? Are you excited by science at the interface of classical AI reasoning and Large Language Models (LLMs)? Would you like to apply your technology to serve operations customers better? Amazon Robotics is looking for a talented Applied Scientist in Neurosymbolic AI. You will innovate on combining language models (LMs) with classical AI reasoning. You will work with a team of scientists and engineers to achieve this. You will publish your results in papers at leading venues in AI. You will be part of a larger team and have the opportunity to work on problems such as: using LMs to generate plans, using AI reasoning to verify plan correctness, learning efficient reasoning strategies, self-improving models. You will work on basic science and on business problems in robotics, automation and fulfillment across our operations. Key job responsibilities In this role you will: • Work closely with other scientists and engineers, and be part of Amazon’s diverse global science community. • Publish your research in top-tier academic venues and hone your presentation skills. • Be inspired by challenges and opportunities to invent new techniques in your area(s) of expertise. A day in the life You'll meet regularly with your technical lead and your team on your ideas, get guidance and feedback, work together on architectures and algorithms, author papers, build AI systems, all with the aim of delivering results for your operations customers. You'll work closely with other scientists to review your plans and results. You'll meet with engineers to implement your ideas at scale. About the team The Veritas team is a science team working at the boundary between language models and classical AI reasoning. We work across on customer problems in fulfillment, automation and robotics. We focus on high quality research science informed by practical problems.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Economist I, POE
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, WA, Seattle

            
        


        
    
Economists in this role partner with business stakeholders to distill complex problems into testable economic questions and generate actionable insights. They collaborate with engineers and scientists to estimate models on large-scale data, design pilots, measure impact, and scale successful prototypes into improved policies and programs. They leverage AI tools to scale economic study for broader business impact. They communicate findings to business leaders, incorporate feedback, and deliver customer-centric solutions at scale.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Data Scientist, ADE Analytics
            

        

        

            
                
                    

                
            

            

            

            

            
                
CA, BC, Vancouver

            
        


        
    
The Alexa Daily Essentials team delivers experiences critical to how customers interact with Alexa as part of daily life. Alexa users engage with our products across experiences connected to Timers, Alarms, Calendars, Food, and News. Our experiences include critical time saving techniques, ad-supported news audio and video, and in-depth kitchen guidance aimed at serving the needs of the family from sunset to sundown. As a Data Scientist on our team, you'll work with complex data, develop statistical methodologies, and provide critical product insights that shape how we build and optimize our solutions. You will work closely with your Analytics and Applied Science teammates. You will build frameworks and mechanisms to scale data solutions across our organization. If you are passionate about redefining how AI can improves everyone's daily life, we’d love to hear from you. Key job responsibilities Problem-Solving - Analyze complex data to identify patterns, inform product decisions, and understand root causes of anomalies. - Develop analysis and modeling approaches to drive product and engineering actions to identify patterns, insights, and understand root causes of anomalies. Your solutions directly improve the customer experience. - Independently work with product partners to identify problems and opportunities. Apply a range of data science techniques and tools to solve these problems. Use data driven insights to inform product development. Work with cross-disciplinary teams to mechanize your solution into scalable and automated frameworks. Data Infrastructure - Build data pipelines, and identify novel data sources to leverage in analytical work - both from within Alexa and from cross Amazon - Acquire data by building the necessary SQL / ETL queries Communication - Excel at communicating complex ideas to technical and non-technical audiences. - Build relationships with stakeholders and counterparts. Work with stakeholders to translate causal insights into actionable recommendations - Force multiply the work of the team with data visualizations, presentations, and/or dashboards to drive awareness and adoption of data assets and product insights - Collaborate with cross-functional teams. Mentor teammates to foster a culture of continuous learning and development




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Senior Applied Scientist, Ad Measurements Science 
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, NY, New York

            
        


        
    
The Ads Measurement Science team in the Measurement, Ad Tech, and Data Science (MADS) team of Amazon Ads serves a centralized role developing solutions for a multitude of performance measurement products. We create solutions which measure the comprehensive impact of advertiser's ad spend, including sales impacts both online and offline and across timescales, and provide actionable insights that enable our advertisers to optimize their media portfolios. We also own the science solutions for AI tools that unlock new insights and automate high-effort customer workflows, such as custom query and report generation based on natural language user requests. We leverage a host of scientific technologies to accomplish this mission, including Generative AI, classical ML, Causal Inference, Natural Language Processing, and Computer Vision. As a Senior Applied Scientist on the team, you will be at the forefront of innovation, developing measurement solutions end-to-end from inception to production. You will set the technical vision and innovate on behalf of our customers. You will propose, design, analyze, and productionize models to provide novel measurement insights to our customers. You will partner with engineering to deploy these solutions into production. You will work with key stakeholders from various business teams to enable advertisers to act upon those metrics. Key job responsibilities * Lead the development of ad measurement models and solutions that address the full spectrum of an advertiser's investment, focusing on scalable and efficient methodologies. * Collaborate closely with cross-functional teams including engineering, product management, and business teams to define and implement measurement solutions. * Use state-of-the-art scientific technologies including Generative AI, Classical Machine Learning, Causal Inference, Natural Language Processing, and Computer Vision to develop state of the art models that measure the impact of ad spend across multiple platforms and timescales. * Drive experimentation and the continuous improvement of ML models through iterative development, testing, and optimization. * Translate complex scientific challenges into clear and impactful solutions for business stakeholders. * Mentor and guide junior scientists, fostering a collaborative and high-performing team culture. * Foster collaborations between scientists to move faster, with broader impact. * Regularly engage with the broader scientific community with presentations, publications, and patents. A day in the life You will solve real-world problems by getting and analyzing large amounts of data, generate business insights and opportunities, design simulations and experiments, and develop statistical and ML models. The team is driven by business needs, which requires collaboration with other Scientists, Engineers, and Product Managers across the advertising organization. You will prepare written and verbal presentations to share insights to audiences of varying levels of technical sophistication. Team video https://advertising.amazon.com/help/G4LNN5YWHP6SM9TJ About the team We are a team of scientists across Applied, Research, Data Science and Economist disciplines. You will work with colleagues with deep expertise in ML, NLP, CV, Gen AI, and Causal Inference with a diverse range of backgrounds. We partner closely with top-notch engineers, product managers, sales leaders, and other scientists with expertise in the ads industry and on building scalable modeling and software solutions.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Applied Scientist, Ads Measurement Science
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, NY, New York

            
        


        
    
The Ads Measurement Science team in the Measurement, Ad Tech, and Data Science (MADS) team of Amazon Ads serves a centralized role developing solutions for a multitude of performance measurement products. We create solutions which measure the comprehensive impact of advertiser's ad spend, including sales impacts both online and offline and across timescales, and provide actionable insights that enable our advertisers to optimize their media portfolios. We also own the science solutions for AI tools that unlock new insights and automate high-effort customer workflows, such as custom query and report generation based on natural language user requests. We leverage a host of scientific technologies to accomplish this mission, including Generative AI, classical ML, Causal Inference, Natural Language Processing, and Computer Vision. As an Applied Scientist on the team, you will lead measurement solutions end-to-end from inception to production. You will propose, design, analyze, and productionize models to provide novel measurement insights to our customers. Key job responsibilities Leverage deep expertise in one or more scientific disciplines to invent solutions to ambiguous ads measurement problems Disambiguate problems to propose clear evaluation frameworks and success criteria Work autonomously and write high quality technical documents Implement a significant portion of critical-path code, and partner with engineers to directly carry solutions into production Partner closely with other scientists to deliver large, multi-faceted technical projects Share and publish works with the broader scientific community through meetings and conferences Communicate clearly to both technical and non-technical audiences Contribute new ideas that shape the direction of the team's work Mentor more junior scientists and participate in the hiring process About the team We are a team of scientists across Applied, Research, Data Science and Economist disciplines. You will work with colleagues with deep expertise in ML, NLP, CV, Gen AI, and Causal Inference with a diverse range of backgrounds. We partner closely with top-notch engineers, product managers, sales leaders, and other scientists with expertise in the ads industry and on building scalable modeling and software solutions.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Applied Scientist, Sales AI
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, WA, Seattle

            
        


        
    
Are you interested in shaping the future of Advertising and B2B Sales? We are a growing team with an exciting AI-first charter and need your passion, innovative thinking, and creativity to help take our products to new heights. Amazon Advertising is one of Amazon's fastest growing and most profitable businesses, responsible for defining and delivering a collection of advertising products that drive discovery and sales. Our products are strategically important to our businesses driving long term growth. We break fresh ground in product and technical innovations every day! Within the Advertising Sales organization, we are building a central AI/ML team and are seeking top Applied Science talent to help us build new, science-backed services that drive success for our customers. Our goal is to transform the way account teams operate by creating actionable insights and recommendations they can share with their advertising accounts, and ingesting Generative AI throughout their end-to-end workflows to improve their work efficiency. As an Applied Scientist on the team, you will bring deep expertise in modeling dynamic systems using statistical methods and deep learning, and in optimizing those systems using reinforcement learning and operations research. You have the scientific and technical skills to build and refine models that can be implemented in production, and you leverage natural language processing and generative AI to enhance their explainability. You will chart new courses with our ad sales support technologies, and you have the communication skills necessary to explain complex technical approaches to a variety of stakeholders and customers. You will be part of a team of fellow scientists and engineers taking iterative approaches to tackle big, long-term problems. You are fluently able to leverage the latest generative AI systems and services to accelerate and improve your work while maintaining high quality in your outputs. Key job responsibilities Scientific Modeling - Conceptualize and lead state-of-the-art research on new Machine Learning and Generative Artificial Intelligence solutions to optimize all aspects of the Ad Sales business - Lead the technical approach for the design and implementation of successful models and algorithms in support of expert cross-functional teams delivering on demanding projects - Run regular A/B experiments, gather data, and perform statistical analysis - Improve the scalability, efficiency and automation of large-scale data analytics, model training, deployment and serving - Publish scientific findings in reports and papers that can be shared internally and externally Product Development Support - Partner with software engineering and product management teams to support product and service development, define success metrics and measurement approaches, and help drive adoption of innovative new features for our services. - Lead requirements gathering sessions with product teams and business stakeholders - Maintain scientific documentation and knowledge for product initiatives Collaboration & Communication - Work closely with software engineers to deliver end-to-end solutions into production - Translate complex scientific findings into actionable business recommendations for stakeholders and senior management - Provide clear, compelling reports and presentations on a regular basis with respect to your models and services - Communicate with internal teams to showcase results and identify best practices. About the team Sales AI is a central science and engineering organization within Amazon Advertising Sales that powers selling motions and account team workflows via state-of-the-art of AI/ML services. Sales AI is investing in a range of sales intelligence models, including the development of advertiser insights, recommendations and Generative AI-powered applications throughout account team workflows.




        

        

        

        
    


    



                    

                
                    

                        

    

    

        
            

                Senior Applied Scientist , Sponsored Products and Brands Ads Response Prediction 
            

        

        

            
                
                    

                
            

            

            

            

            
                
US, NY, New York

            
        


        
    
About Sponsored Products and Brands The Sponsored Products and Brands team at Amazon Ads is re-imagining the advertising landscape through generative AI technologies, revolutionizing how millions of customers discover products and engage with brands across Amazon.com and beyond. We are at the forefront of re-inventing advertising experiences, bridging human creativity with artificial intelligence to transform every aspect of the advertising lifecycle from ad creation and optimization to performance analysis and customer insights. We are a passionate group of innovators dedicated to developing responsible and intelligent AI technologies that balance the needs of advertisers, enhance the shopping experience, and strengthen the marketplace. If you're energized by solving complex challenges and pushing the boundaries of what's possible with AI, join us in shaping the future of advertising. Key job responsibilities As a Machine Learning Applied Scientist, you will: * Conduct deep data analysis to derive insights to the business, and identify gaps and new opportunities * Develop scalable and effective machine-learning models and optimization strategies to solve business problems * Run regular A/B experiments, gather data, and perform statistical analysis * Work closely with software engineers to deliver end-to-end solutions into production * Improve the scalability, efficiency and automation of large-scale data analytics, model training, deployment and serving * Conduct research on new machine-learning modeling and Generative AI solutions to optimize all aspects of Sponsored Products and Brands business About the team The Ad Response Prediction team within Sponsored Products and Brands (SPB) drives personalized shopping experiences for SPB Ads across placements, pages, and devices worldwide. We achieve this through ML and GenAI solutions that include customized shopper response prediction and session-level understanding to optimize every stage of the ad-serving process, from sourcing and bidding to widget discovery and auctions. Our responsibilities include advancing response prediction through model and feature innovations and extending prediction beyond the auction stage to areas such as targeting, sourcing, and bidding.




        

        

        

        
    


    



                    

                
            

        

    

    

        See more jobs
    






        




    

        
            
        
        

            

                
                    

                        

                    

                
            

            
                

                    

                        

  

      
          

              
                  
    
        View from space of a connected network around planet Earth representing the Internet of Things.
    

              
          

      

      

        

          
              
Get more from Amazon Science

          

          
    
Subscribe to our newsletter



        

        
          
        
      

  



                    

                

            
        

    


    

        

        
            
Amazon.com | Conditions of Use | Privacy | © 1996-2026 Amazon.com, Inc. or its affiliates

        
        
            

    
Social