Megan Silvey

Secret analyst habits: - Will rewrite an entire SQL process just because they didn't like the way it was formatted. - Say "it depends" more often than a therapist with a philosophy degree. - Believe "clean data" lives in the same place as unicorns, fairies and leprechauns. - Still genuinely thinks they'll get around to "doing the documentation later."

1,572

108 Comments

Why do entry-level data roles ask for 2+ years of experience? Because it’s not just about technical skills. I rarely hire analysts straight out of school with no work experience. They can code, but they haven’t had time to learn how the business actually works. When people ask me how to break in, I usually say: 1. Get a job in the business. 2. Use your data skills to stand out. 3. Learn how decisions are made, what leaders care about, and how workflows actually function. Good data work supports decisions and fits the messy, human reality of how business teams operate. Think of it as School, Part 2: Learning how your stakeholders think.

216

104 Comments

Secret truths of being an analyst: 1. The file labeled _FINAL is never final. 2. The one request you didn’t design for will be the first question they ask. 3. The stakeholders who say “I don’t need detail” always want drilldowns. 4. No matter how well you tested it, it will break during a live demo. 5. There’s always an unwritten rule missing from the “business rules” you were given.

63

22 Comments

Your data job starts Monday. If you’re still waiting for your first “data” job title, here’s your reminder: You already have a job in data. Don’t wait for someone to hand it to you. Find some data in your current role. Drop it in Excel. Look for something useful. It doesn’t need to be complex or groundbreaking. If you help the business save time, make money, or work smarter, that’s a win. I love it when people bring those examples to interviews.

194

26 Comments

I keep seeing posts about the top skills to be a data analyst and they're all wrong. Excel ranks pretty high. SQL is up there. Python is usually a little lower down. IMO the top skill needed to be an analyst doesn't involve any math or coding - it's curiosity. If you're curious, you can be a good analyst with a pen and a notebook. The best part is when you transition into leadership, it's a skill that continues to serve you well. Tools come and go. Curiosity sticks.

162

48 Comments

The Ledger Analytics API is now open for beta testing! We’ve talked to a lot of actuaries who are excited about fitting Bayesian models to insurance data at scale, but not so excited about wiring up all the infrastructure to make it happen. They want to reliably estimate full posterior distributions on ultimate loss (ratio) predictions, and they want priors that are supported by industry data. That’s why we are opening up our Ledger Analytics API along with a convenient Python interface that makes it easy to get full posterior distributions on ultimate loss ratios with minimal setup. While there are existing tools for modeling insurance data, they lack a few key features that we need to appropriately model insurance-linked securities. We’ve spent years developing our API to meet these needs: (1) Advanced Loss Development: Fit loss development and tail development models directly using our API, ranging from Bayesian versions of traditional actuarial models like Chain Ladder to more advanced Bayesian models developed throughout our ongoing R&D process. Our models leverage informative priors derived from industry data. (2) Time Series Forecasting: Fit Bayesian autoregression and state-space time series models to forecast ultimate losses for future accident and policy periods. In insurance forecasting we often only have dozens of observations, and the ultimates for more recent periods are estimates as opposed to observed data. Our forecasting models use industry-informed priors and account for uncertainty in historical ultimate loss estimates, producing forecasts that are well-calibrated across a wide range of P&C product lines. (3) Open-Source Integrations: Ledger Analytics is heavily integrated with Bermuda, our open-source Python library for insurance loss triangle management and manipulation. Users can also perform model stacking using BayesBlend, our library for performing Bayesian model stacking. (4) Comprehensive Model Documentation: Our documentation includes in-depth tutorials of loss development and forecasting, mathematical specifications of all available models, and details on how each model can be configured.  Coming Soon: Keep an eye out for upcoming features, including loss payment cashflow modeling and ILS deal modeling capabilities! Ready to get started? - Email analytics@ledgerinvesting.com to register and get your API key - Install the package: `pip install ledger-analytics` - Check out the documentation: https://lnkd.in/gZCPthGF

38

2 Comments

🚨 Scam Alert 🚨 I was thrilled when I received an email stating that my job application had been approved for a Remote Data Entry Clerk position. For a moment, I thought I was one step closer to landing an opportunity. But then the red flags started showing: The email asked me to complete a “monetary status check” before even interviewing. The link looked suspicious. When I tried calling the number listed on their website, there was no response. This is a reminder that if something feels off during the hiring process, trust your instincts. A real employer will never ask for personal financial details before hiring. To anyone job hunting right now: be careful, stay alert, and don’t let scammers take advantage of your excitement and hard work. Let’s keep sharing these experiences so others don’t fall into the same trap. #JobSearch #ScamAlert #StaySafe #CareerTips

1,478

155 Comments

🧠 𝗧𝗵𝗲 𝗥𝗲𝘀𝘂𝗺𝗲 𝗠𝗶𝘀𝘁𝗮𝗸𝗲 𝗧𝗵𝗮𝘁 𝗦𝗶𝗹𝗲𝗻𝘁𝗹𝘆 𝗛𝗼𝗹𝗱𝘀 𝗕𝗮𝗰𝗸 𝗦𝗼 𝗠𝗮𝗻𝘆 𝗔𝘀𝗽𝗶𝗿𝗶𝗻𝗴 𝗗𝗮𝘁𝗮 𝗦𝗰𝗶𝗲𝗻𝘁𝗶𝘀𝘁𝘀 Every week, I see dozens of resumes from entry-level data professionals — bootcamp grads, fresh grads, career switchers. Here’s the mistake I see most often, and it rarely gets called out: 𝗣𝗲𝗼𝗽𝗹𝗲 𝗳𝗼𝗰𝘂𝘀 𝗼𝗻 𝘁𝗵𝗲 𝘁𝗼𝗼𝗹𝘀 𝘁𝗵𝗲𝘆 𝘂𝘀𝗲𝗱 — 𝗻𝗼𝘁 𝘁𝗵𝗲 𝗽𝗿𝗼𝗯𝗹𝗲𝗺𝘀 𝘁𝗵𝗲𝘆 𝘀𝗼𝗹𝘃𝗲𝗱. Let me show you what that looks like: ❌ “𝗕𝘂𝗶𝗹𝘁 𝗮 𝗰𝗹𝗮𝘀𝘀𝗶𝗳𝗶𝗰𝗮𝘁𝗶𝗼𝗻 𝗺𝗼𝗱𝗲𝗹 𝘂𝘀𝗶𝗻𝗴 𝗫𝗚𝗕𝗼𝗼𝘀𝘁” → Okay, but why? What was the context? What did it achieve? ✅ “Increased customer churn prediction accuracy by 14% using XGBoost, helping the team proactively retain at-risk users” → Now I understand what you did, why you did it, and what impact it had. In data science, your job isn’t to use algorithms — it’s to drive outcomes through thoughtful application of data. You’re not being hired because you know Python. You’re being hired because you know how to think with data. And that shows up in your resume like this: 🔍 𝟯 𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀 𝘁𝗼 𝗔𝘀𝗸 𝗬𝗼𝘂𝗿𝘀𝗲𝗹𝗳 𝗪𝗵𝗲𝗻 𝗪𝗿𝗶𝘁𝗶𝗻𝗴 𝗮 𝗗𝗦 𝗥𝗲𝘀𝘂𝗺𝗲: 1.What was the business or real-world problem I was trying to solve? 2.How did I approach it, and why did I choose that method? 3.What measurable result or insight came from my work? If you can answer these clearly, your resume will stand out — no matter your background. 💬 If you're currently reworking your resume, I’m happy to review 1-2 lines in the comments. Sometimes one tweak can make all the difference. 𝐄𝐱𝐭𝐫𝐚 𝐍𝐨𝐭𝐞: If you're eager to master the full spectrum of 𝐝𝐚𝐭𝐚 𝐬𝐜𝐢𝐞𝐧𝐜𝐞, 𝐦𝐚𝐜𝐡𝐢𝐧𝐞 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠, 𝐚𝐧𝐝 𝐀𝐈, I highly recommend the 𝐅𝐮𝐥𝐥 𝐒𝐭𝐚𝐜𝐤 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐜𝐞 & 𝐀𝐈 𝐁𝐨𝐨𝐭𝐜𝐚𝐦𝐩 from 𝐓𝐞𝐜𝐡𝐕𝐢𝐝𝐯𝐚𝐧. 𝐄𝐱𝐩𝐥𝐨𝐫𝐞 𝐭𝐡𝐞 𝐜𝐨𝐮𝐫𝐬𝐞 𝐡𝐞𝐫𝐞:-https://lnkd.in/dED2TBc9 𝐇𝐢𝐠𝐡𝐥𝐢𝐠𝐡𝐭𝐬 ✅ Full Stack Data Science, Machine Learning, and AI programs ✅ Personalized mentorship from industry experts ✅ Live, interactive classes with 24/7 support ✅ Hands-on projects to boost your job readiness 🎯 Start today and chart your path to becoming a Data Scientist! Ready to make an impact? 🔥

65

23 Comments

I just analyzed 1,000+ job postings for data roles in 2026. Here's what I found. (The market has completely shifted.) Companies aren't hiring the way they used to. Welcome to the "precision hiring" era. Fewer roles, higher expectations, bigger paychecks for specialists. I used Superagent, a newly launched AI product from Airtable, to research this. Right away, multiple specialized AI agents worked simultaneously to pull from sources like LinkedIn Jobs, Glassdoor, and industry reports, to deliver an interactive 𝐒𝐮𝐩𝐞𝐫 𝐑𝐞𝐩𝐨𝐫𝐭 with salary benchmarks, skill trends, and hiring patterns. 𝐇𝐞𝐫𝐞'𝐬 𝐰𝐡𝐚𝐭 𝐭𝐡𝐞 𝐝𝐚𝐭𝐚 𝐬𝐡𝐨𝐰𝐬: 𝟐𝟎𝟐𝟔 𝐒𝐚𝐥𝐚𝐫𝐲 𝐁𝐞𝐧𝐜𝐡𝐦𝐚𝐫𝐤𝐬 (𝐌𝐢𝐝𝐩𝐨𝐢𝐧𝐭): ↳ Data Analyst: $117,250 (+3.3% YoY) ↳ Data Scientist: $153,750 (+4.1% YoY) ↳ Data Engineer: ~$145,000 ↳ ML Engineer: $170,750 (+4.4% YoY) 𝐓𝐡𝐞 𝐒𝐤𝐢𝐥𝐥𝐬 𝐒𝐡𝐢𝐟𝐭: ↳ 60% of Data Scientist postings now require AI skills ↳ LLM experience is the #1 demanded AI skill ↳ Python + SQL are still non-negotiable ↳ Cloud platforms (AWS, Snowflake) showing up in 41% of DS roles 𝐁𝐢𝐠𝐠𝐞𝐬𝐭 𝐈𝐧𝐬𝐢𝐠𝐡𝐭𝐬: → ML Engineer is the #1 fastest-growing role in the US (per LinkedIn) → Entry-level roles are shrinking. Companies want production-ready skills. → AI-skilled professionals command 5-12% salary premiums → The market isn't saturated, it's "misaligned". Employers want judgment, not just theory. 𝐌𝐲 �� 𝐜𝐞𝐧𝐭𝐬: The days of learning Python and landing a data job are fading. 2026 rewards specialists who can ship end-to-end, work with AI tools, and make decisions under uncertainty. If you're job hunting this year, focus on production experience over certifications. If you want to check out the full Super Report I generated, comment “𝐑𝐞𝐩𝐨𝐫𝐭” below and I'll send it to you. The full report includes interactive salary charts, geographic breakdowns, and top hiring companies. All with citations you can actually verify 𝐓𝐫𝐲 𝐢𝐭: https://lnkd.in/dBMd5wTU Use my code "VENKATA2FREE" to get 2 months free for Superagent! Which role are you targeting in 2026? 👇 ♻️ Repost to help someone plan their data career this year #SuperagentPartner Airtable Superagent

175

87 Comments