Listen on
Overview
Host Ross Katz speaks with Miguel Cacho Soblechero, Data Scientist at Prolaio. Fewer than 10% of drugs entering Phase 1 clinical trials ever reach approval, representing billions in lost investment and years of scientific effort. This staggering failure rate stems partly from clinical trials operating on historical experience rather than data-driven insights. While trials generate massive datasets, many companies struggle to use this information proactively to improve patient recruitment, reduce dropouts, and enhance operational oversight.
This episode confronts that challenge head-on. Miguel, whose background spans medical device engineering and advanced data science, outlines how data science teams are fundamentally changing clinical trial execution. He details the practical steps to move trials from a reactive, experience-based model to a proactive, data-informed system.
The discussion covers everything from optimizing protocol design using electronic health records to deploying predictive models for patient engagement, and smart alerting systems for trial monitoring. Miguel also provides a clear roadmap for smaller biotech companies to build data maturity, emphasizing foundational steps before pursuing advanced AI, ultimately positioning data science as a critical tool for reducing risk and accelerating drug development.
Key Takeaways
Clinical Trial Dropouts Drain Resources; Data Science Can Stop It.
Patient dropout rates in clinical trials render collected data unusable and represent a significant financial loss. Applying engagement models from e-commerce, data science teams can predict which participants are likely to churn. Proactive interventions based on these predictions safeguard trial integrity, preserve valuable data, and reduce overall costs.
Prioritize Curated Internal Data Over Noisy External Records.
While aggregated electronic health records offer scale, their billing-centric design makes them inherently noisy and challenging for data science. Long-term strategic value lies in meticulously curating internal datasets from structured clinical trials. This internal data becomes a unique asset, allowing organizations to conduct powerful meta-analyses across studies, identify high-performing sites, and understand trial success factors at a deeper level.
Build AI Maturity Incrementally: The Data Scientist’s Staircase.
For biotech companies, especially smaller ones, expecting immediate ‘transformer model’ results from AI is unrealistic and costly. True value comes from climbing a ‘data scientist’s staircase’: starting with reliable data warehousing and exploration, moving to simple visualizations and rules-based insights, and then gradually incorporating advanced machine learning. This iterative approach ensures maintainability, user understanding, and sustained organizational data literacy.
Data Shifts Clinical Operations from Reactive to Proactive.
Moving clinical trials from reactive problem-solving to proactive intervention is crucial for safety and efficiency. Data science-driven alerting systems can flag issues like patient attendance drops or unexpected adverse events early. Presenting objective data breaks down internal ‘political egos,’ allowing teams to focus on facts and collaborate on solutions rather than engaging in finger-pointing, ultimately making trial execution more efficient and improving patient safety.
Related: CorrDyn specializes in helping biotech and life sciences companies build strong data engineering foundations. We often begin with a data maturity assessment to chart a realistic path for organizations to access their data’s value.
Full Transcript
Jason: Welcome to Data in Biotech, a podcast from CorrDyn where we explore how companies leverage data to drive innovation in life sciences. Every two weeks we sit down with an expert from the world of biotechnology to understand how they’re using data science to solve technical challenges, streamline operations and further innovation in their business. This week, we’re excited to be joined by Miguel Cacho Soblechero, a data scientist helping biotech and pharma companies make the most out of their data. During this interview, our host Ross Katz speaks with Miguel about his transition from academia into the field of data science and biotechnology. They also cover the exact step-by-step process in which a data science team will support a clinical trial, from the moment it’s decided what’s going to be tested and on whom, all the way through to generating data insights and getting them in front of the people who need them most. Then they unpack the realities of what’s possible with AI and machine learning that could ultimately lead to better drugs and healthcare. Here we go.
Ross Katz: Miguel Cacho Soblechero, welcome to the Data in Biotech podcast!
Miguel Cacho Soblechero: Thank you so much, Ross. Pleasure to be here today.
Ross Katz: Great to have you. To kick us off, could you give me an overview of your career to date?
Miguel Cacho Soblechero: Sure, absolutely. My career has been a little bit unconventional as probably any data scientist career. My background is definitely not on data science. I come from a purely engineering background. I studied back in Spain. I focused on medical devices. So programming these little devices that they put you on hospitals to know what you have and how to diagnose you better. I started like that and then slowly and steadily I’ve become more and more passionate about how those three things, how engineering, data, and healthcare can be brought together to make our lives better. That took me through my master’s, my PhD at Imperial, and now I’m across the pond. I’m right now based in Boston, helping biotech and pharma companies to make the most out of their data.
Ross Katz: That’s great. I know you’ve experienced and thought a lot about the transition from PhD into industry and data science in particular. Can you share some of the lessons learned you have from making that transition?
Miguel Cacho Soblechero: Absolutely. Probably everyone that has done a PhD on something related with data, their main focus during the PhD has been get the best model, get the best accuracy, get the most out of the data. And that is still true when you make the transition to industry, but I think it’s with a twist. You need to start counting the human factor. And that’s something that was probably the most interesting part of my transition from the PhD to academia is- data is important, but how the data is going to be used by your users is probably the key question that you need to start with and then roll back to your model, your data processing, everything that you can do with data.
Ross Katz: Thinking through not just the model itself, but how that model will be applied and all of the constraints and the requirements that come out of the different people that are going to be touching the outputs of the model.
Miguel Cacho Soblechero: Absolutely. For me, the biggest learning is to think about the process in which my tools are going to be used. Not only about the user, there is a big focus and if you read any entrepreneurial book it’s all about, ‘Oh, just be user focused,’ and user focus is critical. But you need to understand how is the life of your user? How are you going to change their process? How are you going to change their life? And once you understand how are you going to change their lives with data, you can roll back and make something useful for them.
Ross Katz: Talk to me about why you do what you do. What motivates you to apply data science in this field?
Miguel Cacho Soblechero: What first got me into healthcare is the immediate impact that you can have in people’s lives. First of all, the novelty. I think it’s a field that is still starting, both from the engineering perspective and from the data science perspective. There’s a lot more that we can know with the data and with the engineering that we have. That opportunity is the one that got me into the field. What maintained me into the field is the excitement. It’s the impact that you can have on people’s life and the opportunities that you have with the data that we have right now. Right now we’re living on a world in which, thank God we’re getting more and more data on more various sources in the world of healthcare. We have the genomics revolution with the genome project that has created an explosion of data. We have on clinical trials we will talk in a minute that essentially is filled with data. And that’s an untapped opportunity. No one right now is using that data and I think right now is the right time to be in the right place and that right place is healthcare and data.
Ross Katz: As you just mentioned, you’ve had the opportunity to work with the data ecosystem surrounding clinical trials. Can you give us a little background on your experience with supporting clinical trials and then give us an overview of the data that you’re using, the sorts of problems that you solve?
Miguel Cacho Soblechero: Sure, absolutely. For the audience that might not be familiar with what exactly a clinical trial is or how it’s structured, clinical trials are essentially the backbone of how pharma and biotech industry works and is the most objective way that we have to test whether something that we are going to give to our patients actually work. And the way we do it is by creating different groups with essentially different treatments, giving them whatever we are assessing, and discovering whether that treatment actually helps the patient or it doesn’t, or it has some side effects that we didn’t initially thought. The opportunity here is massive. The reason why is because the amount of data that is generated during these clinical trials is humongous throughout the entire process. From the very beginning of the project, from the very beginning of the trial, we start with a protocol design, which you can think it as the instruction manual, the user guide of the clinical trial. And then throughout the entire process, from the moment that we decide what are we going to test, where are we going to test it, who are we going to test it with, and how are we going to measure it that this is safe and it’s actually doing what we promise, we are generating data and we have the opportunity to put that data in front of the people that need it the most, the patients, the people that are working on the day-to-day basis on the clinical trials.
Ross Katz: That makes sense. What would help to make it concrete would be to just talk through a single drug that’s being ushered through clinical trials and just talking about how, in your experience, a data science team supports a clinical trial today.
Miguel Cacho Soblechero: The first time the data science team gets involved is during the protocol design. Protocol design is the backbone. It’s the user manual. And the way user manuals or protocol designs are done now are either by past experience or knowledge, but it’s really hard to assess the implications of the decisions. You are going to decide which patients are suitable for your clinical trial, but you have no way of knowing the impact of your decisions. And that’s where the data science team, for example, can come in. They understand from EHR, which is electronic health record, essentially the data that we all generate when we go to a doctor, and of course, just to clarify, this data is always anonymized so we won’t be able to tell who is behind the data. But you can aggregate that data and understand, if I change the criteria, how many patients am I actually targeting? How many patients am I able to recruit on this clinical trial? We are moving to this kind of experience-based approach to a data-driven culture that can really help pin down what are the implications, what are the consequences of the decisions that we take. And that has a knock-on effect. If you nail the inclusion/exclusion criteria, the next step, which is trial initiation, is going to be a lot easier. Trial initiation is from the moment that we have the user manual to the moment that we start getting patients inside the clinic and start giving them either the medication, the algorithm, or the device that we are trying to evaluate. And in that trial initiation, data science, again, is critical. Why? Because the same way we can detect what are the patients that are going to be affected by this drug, we can find which providers can help us recruiting those patients. You can start seeing how all the data that is generated for other purposes can start getting merged together and help us throughout the clinical trial.
Ross Katz: To make sure that I’m understanding correctly, in the proposal stage, you’re understanding what the universe of available patients is like, what the target population size is so that you’re making sure that there’s enough patients out there that you can actually run the clinical trial. And then in the next phase, when you’re trying to recruit providers to help you recruit patients for the clinical trial, can you walk me through what the input is that the data science team gives in that sort of situation?
Miguel Cacho Soblechero: Absolutely, Ross. Just to take a step back, I’m going to talk about how it is done now and how data can help. Because I think right now a lot of the main industry is run on experience. You have a drug for a specific condition and probably in your team you’re going to have a physician, a doctor, that is an specialist on this condition. And that doctor will have a network of 10, 15, 20 colleagues that are also doing research and that run clinical trials. That is a really small pool of people. Think about it. Probably people in Boston know people in Boston, or in New York, or maybe they go to Miami on holiday and they know some people in Florida. But that really reduces both the diversity of the people running the clinical trial, the diversity of our population, and the amount of patients that we can reach. Data is coming to change completely that. Because now we can expand that knowledge, that experience that our investigator have, which is really important, it will still be crucial. But we can expand it with the data that comes from our medical records and the information that we get from what each provider is doing. And we can reach people that might be really interested on doing a clinical trial but they are untapped because they might be in remote regions or because they simply not moving in the circles that classical biotech and pharma are moving on. That is incredibly important. And I think that is crucial, as I mentioned, to increase dramatically the amount of patients that we can address and also change the culture, which is really important, to move from experience-driven to a data-driven culture inside our organizations.
Ross Katz: It’s from a population sizing problem to more of a targeting problem in that second phase where, in the past you would have to go through the providers because that was the only way to access the patients. But now you’re able to identify where these patients might be located at a more granular level so that you can fill the clinical trial with a more diverse population. After the clinical trial is operating, you’ve recruited the patients into the trial, what role does the data science team play next?
Miguel Cacho Soblechero: Massive. Once you have the participants, the first thing that you need to make sure is that those participants remain engaged. This is a massive problem in clinical trials. Anyone that you ask, they’ll probably tell you the same. There’s a big dropout of participants and participants leaving the clinical trial essentially make their data useless. And we can leverage tools that have been there for a long time. Amazon does this really well, which is keeping your clients engaged. This microphone was bought, the one that you are seeing here, because Amazon keep recommending me microphones and keep me engaged. We can use the same ideas, the same ML mechanism, to keep our patient engaged. By understanding the data that they generate during the clinical trials and in digital tools, we can predict which participants are likely to churn and we’ll be able to make something about it and keep them motivated. That is probably the first step. The second step is about monitoring the execution. You can imagine that running a small organization is hard. If you start making this globally, is really complicated because you have cultural differences, you have misunderstandings of the protocol, et cetera. This is where data is really important. Just so we understand, every time our patient goes into a clinic for a clinical trial, they need to fill a whole bunch of form saying who they are, what drug are they taking, et cetera. And we get back that data with their lab results, their demographic, et cetera. And we can process it and make sense of it and help this monitoring to be smart. If we see, for instance, that a specific hospital is lagging behind and suddenly the patients stop coming, we can make something about it before it’s too late. If we see that a participant is slowly generating more adverse events than what you were expecting from the average, let’s do something about it. Let’s bring that patient to the clinic and say, ‘Hey, what’s going on?’ Those approaches help everyone throughout the clinical trial.
Ross Katz: Just to make sure that I’m understanding correctly. In the monitoring phase, there’s the operational level monitoring of noticing a dropout in engagement from your patient population and then there’s also the medical level of monitoring. At the operational level, what are the types of interventions that you would do if you discover that patients are no longer coming to the clinic, for example? How do people act on the data at that point?
Miguel Cacho Soblechero: The approach that we took on our previous organization and what we were discussing is essentially an alerting system. You can think this as some kind of layered systems. The site talks with an intermediary that then talk with the organization that then- there is a lot of people in the middle. What we want is to actually reverse this. The people that are on the pharma know in advance from the data what is happening and they can question directly, ‘Hey, what’s happening? Why are the participants missing X, Y, or Z intervention?’ And that triggers the entire conversation. It might be simply that they misinterpreted the protocol, but that’s fine. The important thing is that we are shifting the culture again. Right now, running on a reactive culture. You do something when something happens. We want to move to a proactive culture in which you get the data and you can react, you can proactively take actions with the early signs. That shift in culture can help everyone throughout the clinical trial. It can help the patient increasing their safety. It can help the site running faster the trial and it can help the pharma saving costs which potentially will reduce the risk and help the patient at the end. And it’s not only that they bring it to the table, but they bring facts to the table. Because sometimes teams fight, and that’s the reality. What you want is to bring data to break down those fights, break down those political egos and saying, ‘No, I’m doing it right. No, I’m doing it wrong,’ and actually focus on the facts and how we can do things better. Instead of finger-pointing to each other, finger-point to the data and say, ‘What can we do about this?’
Ross Katz: It sounds like for the kinds of problems that happen in clinical trials that the EHR data is critical to the entire process because you’re trying to understand what the patient population looks like, what are the medical challenges that they’re facing that this drug might be able to help with.
Miguel Cacho Soblechero: I think EHR are great because of their scale, but also that brings a lot of noise. Because these datasets are not intended to be used this way. These datasets are for billing. It was created for insurance companies to understand what was happening on the clinic and how we can charge for it. And we are using it for a completely different purpose. Although they have a lot of opportunities, they are incredibly noisy and their results need to be always taken with a pinch of salt. That’s why I think although EHR are important, being able to have your own dataset that is internally curated and internally generated is incredibly valuable. And I think this is where bigger pharmas might have the upper hand in terms of data. On my previous company at IQVIA, we had one or two clinical trials running at the same time, so you can imagine that although each clinical trial generates a lot of data, the diversity of that data was really little. Imagine someone like a Roche, an AstraZeneca that has run thousands and thousands of clinical trials in a really structured way. That’s where they really have a big opportunity to understand not only at the patient level but at the study level. Aggregate all the studies that you have run so far and be able to distinguish what is a normal study and what is not. What studies went well and which one not. Which specific hospitals are known for running good clinical trials and which ones need more attention from us. The quality of the internal data that you can acquire is critical, not only on the short term but as a long-term perspective. As you scale, as you grow, that is going to be one of your biggest assets.
Ross Katz: Is that an element of how the clinical trials are designed as well? That you’re thinking through not just what are the pieces of data that we want about this particular clinical trial, but what are the pieces of data that we want to gather about all the clinical trials so that when we go back to do that analysis that you’re talking about, that meta-analysis of clinical data that you have the ability to do that analysis?
Miguel Cacho Soblechero: I wish the answer was yes, Ross. In my experience, in general at my previous company, it was mixed. All these tools are not thought for data science, are thought for clinical trial monitoring. And I think this is why I come back to the cultural shift. It is up to the data science to liaise with clinical operations, which is the team that usually runs the studies, to be able to convince them, ‘Hey, we need to acquire this data and make it suitable for our future needs.’ That conversation I think needs to happen top-down because these are completely different cultures. Data scientists are known for ‘move fast and break things,’ quoting a little bit Mr. Mark Zuckerberg. Whereas clinical trials, you cannot run fast and break things because those things are humans. Your worry is safety. Those are two different cultures that need to work together for the good and the long term. That’s one of the biggest challenge that data science teams on clinical trials probably have to overcome: make this data not only useful for the clinical trial, but useful for the future of data science.
Ross Katz: Back to the smaller biotech and pharma companies, I’m interested in the EHR type of analysis. Because you say it’s very noisy and it makes sense that it’s been designed for billing purposes, but I’m really interested in how you get access to the data as an emerging biotech company and then what are the types of methods that you’re applying to these text-based, heavily unstructured datasets that have been designed for a completely different purpose?
Miguel Cacho Soblechero: If you are a small biotech or pharma, what I would suggest always is to start small. There are a lot of datasets out there that can help you. And there are a lot of providers. You have Optum, Flatiron. Those companies have been aggregating EHR data for a while. What I would suggest to this kind of companies is focus your effort on one killer application and start climbing what I would call the data science staircase. Start getting the data in your warehouse. That’s going to take you more time than you think. And start building the infrastructure that allows you to explore the data. Once you’re comfortable with that, start thinking not on a complicated AI mechanisms. It’s like, how this data, how this simple visualization can help my users? And then start climbing that ladder by creating simple rules. We started with simple rules, starting creating cohorts that can help our physicians to make sense of what we mentioned before of the inclusion/exclusion criteria. And then as you progress, you can start looking at embeddings, how each of these EHR cluster together. There is huge opportunities. But I would advise, especially for small biotechs, although right now everything is about AI, hold your horses and start climbing that ladder because it is a ladder. You cannot jump any step.
Ross Katz: If I’m understanding you correctly about the maturity model, at first you’re just getting the EHR records that you need for a particular drug that’s under development or a particular trial that you’re preparing to run. And then you’re exploring the data and using rules-based approaches to segment the population to give people an idea of how many people are in different categories that you might want to target for a trial, for example. And then later on you’re thinking about the more advanced NLP methods of training embeddings based on this data, clustering them together, trying to do more with the data so that you’re not relying on such blunt objects as keywords and rules.
Miguel Cacho Soblechero: Absolutely. That’s where the data science leadership is key. Because people like me that just finished the PhD and went for the first time in pharma, they will think that the best idea is to run a Transformer and create a massive architecture. Reality is sometimes the simpler the solution, the better. Because it will be, first of all, easy to maintain, easier to deploy, and easier to communicate. The people that are going to use it need to understand what are they looking at. Communicating how data can be used, what usually is called data literacy, is really important because of the cultural shift that we’re seeing in pharma. It’s really important to establish that dialogue and changing the direction of the dialogue. First, it’s going to be the data science communicating the clinical operations or any other team what are the opportunities. And I think the real success is when that relationship get inverted. When the other team, clinical operations, your stakeholders, understands both the opportunities but also the limitations and can come up with ideas of, ‘Hey, if I’m able to see this specific metric, that will save me a lot of time and a lot of pain. Can you do it based on the data?’ So the ideas are, first of all, focused on the data, focused on what is possible, and focused on what can help them.
Ross Katz: That makes a lot of sense. What you’re talking about mirrors the data science journey in lots of industries, is that you put the data in people’s hands and you show them what’s possible and that creates the opportunity to ask questions of what else might be possible and how time can be saved and what new insights can be generated and that dialogue between the data science team and the domain experts is critical to making that happen.
Miguel Cacho Soblechero: We are in a really difficult environment because the expectations are over the roof. If you type ‘opportunities for AI in healthcare’, they are sometimes not tied to reality. There are like, ‘Boom! ChatGPT is going to replace completely all the clinical operations and clinical trials are going to be run by a bot.’ We might get there. I’m not saying that’s not the possibility, but each organization has its times, has its tempo, and needs to understand how to get there.
Ross Katz: That makes a lot of sense. And also in the diversity of companies that are out there, the ecosystem of biotech and pharma, it seems likely that some of the larger organizations will get there sooner. But that doesn’t mean that there isn’t a lot of value to this journey that you’re talking about for some of the smaller biotech and pharma companies as well.
Miguel Cacho Soblechero: I think AI is probably the biggest opportunity for small organizations, especially when we are talking about clinical trials. Clinical trials are probably the biggest risk factor of any pharma company. If we are able to mitigate that risk factor, if we are able to essentially reduce the failure rate across entire chain- we’ve been only talking about clinical trials but we can talk about drug discovery, if we are able to create better drugs, understand those drugs better from the inception, if we can reduce the cost of running a clinical trial and the time that it takes, that’s going to have a knock-on effect on price and on risk. And that will allow smaller companies to have better chances to have drugs on the market. That is why AI are going to be the big equalizer when it comes to biotech and pharma.
Ross Katz: That’s really interesting. What do you think are some of the barriers that are in place that prevent smaller biotech companies from climbing that ladder, from getting to the world that you’re talking about?
Miguel Cacho Soblechero: Let me give you a number. Only 10% of the drugs that start on phase one make it into approval. And those are 10-plus years of work on a 10% chance. That is a huge barrier and a huge risk. Better understanding the biology and finding better ways of modeling it, it’s going to take that number up and hopefully soon we’ll have more and more smaller biotechs that have new drug and can explode, like the example of Moderna. They started with the COVID vaccine and thanks to that big medicine, now they are able to expand their catalog and they’ve been talking about cancer vaccine, et cetera. I hope that we will see more of those examples in the near future.
Ross Katz: It sounds like the biggest opportunity is in increasing the knowledge and the insightfulness of the core science, of the core biological understanding that’s driving the mechanism by which the drug does what it’s designed to do, and then all of the operational improvements that you’re talking about fall out from there.
Miguel Cacho Soblechero: Exactly. Exactly.
Ross Katz: What I heard you talk about earlier is that for some of the people that have that domain knowledge and potential to do some of the intense machine learning modeling, there’s this inclination to look at a problem and immediately want to train a Transformer model. But a lot of the value comes from the low-hanging fruit of building out…
Miguel Cacho Soblechero: You need to start with baby steps, start with those low-hanging fruits, and start adding complexity. If you start from the top, that’s not going to work. Because at the end of the day, your team at the very beginning are going to be prophets. Communicators of data that will convince the rest of the organization that the data that they are offering is the right data and it’s useful for them. You either have someone that is specialized on communication, that can be a product owner, or you can have a data scientist that has that ability of coordinating teams while making sense of the data.
Ross Katz: That makes a lot of sense. As we look towards the future, five years from now, 10 years from now, what do you foresee as the next advancements that we’re going to see in utilizing AI/machine learning to revolutionize clinical trial processes and patient outcomes?
Miguel Cacho Soblechero: The biggest one, we already seeing it, which is the inclusion of wearables. To me it sounds obvious because we all have smartwatches that control our heart rate all the time, but they’re rarely connected to the clinical trial. There’re organizations that are moving towards that direction and it’s incredibly important because let’s be clear, going to the hospital three times a week for a clinical trial is not pleasant. It’s time-consuming. If you can be monitored your condition from home, that will increase patient compliance massively. And also we will be able to better understand what are the effect of the drug on the body. You are not getting timestamps, you’re actually getting a time series. You’re actually getting the entire history. That’s going to be incredibly important. The second, of course, is generative AI. I know that it sounds a bit cheesy because everyone is talking about it, but I truly believe that generative AI has a immense power when it comes to helping clinical trials. As I mentioned, from the very beginning, from the protocol design, use generative AI to understand what are the things that you put, what are the implication, what are the holes that you might not be looking at, to helping insights through a chatbot to do better the clinical trial, to better understand how the protocol is, that can have massive impact. And then moving towards digital trials and digital twins, so understand the peculiarities of your participants and leverage that on a clinical trial, that can change completely the way we treat patients, the way we run clinical trials, and the way drugs are delivered and pharma works. Those three trends are going to be key to keep an eye on for the next 10 years.
Ross Katz: Having tread this path of entering this industry from your PhD into a data science role, if there’s somebody who’s really interested in moving into data science roles in biotech and pharma, what are the ways that you would encourage people to gain the skills that they need, gain the perspectives that they need to enter this field?
Miguel Cacho Soblechero: My first advice is to don’t be afraid to reach out. Although you see someone that is Chief Data Officer and you’re like, ‘Oh my God, this person is unreachable.’ The reality is that that person very likely has gone through the same journey and most of these people are willing to share their experiences. My first job, the way I got in was simply sending cold messages to people on LinkedIn, asking for advice for someone like me that has finished a PhD and have followed the same journey. We all love to help the new generation and I think not being afraid of asking for advice and find the right mentor for you is critical. From the technical standpoint, there is loads of resources. Don’t get trapped only on the theoretical frameworks, don’t think that you need to know every single algorithm. Definitely they are important, but having a portfolio of projects to show, ‘Hey, I’m interested on these problems and I have done this analysis on this dataset and I found these insights.’ That’s what people are looking for. That combination of a theoretical background, but a practical experience with data, making sense of the data and extracting insights are the fastest way forward. And it will help you through the interview process and your career. Having stories to tell about the things that you have achieved is critical to find the next opportunity.
Ross Katz: That makes a lot of sense. As we bring this conversation to a close, anything you’ve been reading or thinking about lately that you would recommend to our listeners as resources that would help them to get more acquainted with this data science for clinical trials topic?
Miguel Cacho Soblechero: That question is really timely. I have just finished a really short book that is quite funny. It’s called ‘Statistics Done Wrong.’ It talks a little bit exactly on how you need a solid theoretical foundation and you need to understand the basis of it to really understand how to develop models, how to develop statistics, and draw the right conclusions both data science of clinical trials and in life in general. I would definitely recommend people to check out that book. It’s a light read, it’s quite funny and it’s really, really insightful.
Ross Katz: Well, Miguel, it’s been a real pleasure to talk with you today, really appreciate the time, and look forward to connecting down the line.
Miguel Cacho Soblechero: Absolutely, Ross. It has been a pleasure. I really enjoyed this conversation and we’ll stay in touch.
Jason: Are you a biotechnology company looking to unlock the potential of your business data? CorrDyn can help. We’re an enterprise data specialist that helps companies working in life sciences make smarter strategic decisions. From developing the right data strategy that starts with our data maturity assessment to building and delivering bespoke technical solutions, we are equipped to tackle the most complex data challenges. We have partnered with dozens of high-growth organizations, from manufacturers of custom oligonucleotides to molecular diagnostic companies, to achieve data competence. Whether you need to supplement existing technology teams with specialist expertise or launch a data program that lays the groundwork for future internal hires, you can partner with CorrDyn to unlock the potential of your business data today. Simply visit connect.corrdyn.com/biotech to learn more. Now, back to the show. And that’s it for this episode of Data in Biotech. If you enjoyed the episode, please subscribe, rate, or leave a review in your podcast platform of choice. See you next time.
