DeepSeek vs Gemini: Which AI Wins for Recruiters?
DeepSeek's reasoning skills are making waves, but how does it stack up against Google's Gemini for recruitment? Let's explore the pros, cons, and privacy concerns.
You know how it is in our field – we're always on the lookout for that next big thing, that tool that's going to make our lives easier and help us find the perfect candidates faster. A few days ago, I saw a post about DeepSeek that really caught my eye.
Glen Cathey, who's always sharing interesting content when it comes to recruitment (seriously, if you're not following him, you should be!), shared a video about this new AI on the block. It was a video showing off some DeepSeek reasoning skills for creating Boolean searches.
But here's the thing – it got me thinking. Is DeepSeek really that revolutionary? I mean, we already have access to some pretty powerful AI tools ourselves, right? Specifically, I'm talking about Google's Gemini models, including that intriguing "Gemini 2.0 Flash Thinking Experimental" version.
It seemed like they were capable of similar things. So, before we all jump on the DeepSeek bandwagon, I thought it'd be worth taking a closer look how these two AI are thinking about the tasks.
Lets dive into a head-to-head comparison of DeepSeek and Gemini, specifically from a recruiter's perspective, and we will use the same prompt that Glen used so you can see for yourself the results and reasoning process.
I'll address some serious questions about data privacy that you absolutely need to be aware of. Let's face it: In our line of work, protecting candidate information is just as important as finding the right talent.
DeepSeek: The New Kid on the Block
Alright, so I've mentioned DeepSeek a couple of times now, but what exactly is it? DeepSeek is built on some seriously advanced technology, and it's coming out of China, a country that's quickly becoming a major player in the AI race.
DeepSeek is an open-source large language model (LLM), that aims to provide competitive AI solutions at a significantly lower cost than its Western counterparts, particularly OpenAI.
Now, what got everyone so excited about DeepSeek? It all comes down to its reasoning abilities. In that video Glen shared, DeepSeek was shown tackling complex problems (in this case, a Boolean search string) and coming up with logical solutions—the kind of stuff that recruiters do every day.
It’s pretty refreshing to see how AI approaches the problem, especially since we didn’t see this with ChatGPT.
Now, it's important to remember that DeepSeek is still relatively new. But the potential is definitely there. But, and this is a big "but," we can't ignore the elephant in the room, which we'll get to a bit later – the crucial question of data privacy.
I personally wouldn’t suggest using DeepSeek via web access for anything because of privacy concerns. Take a look at DeepSeek’s terms and conditions—it’s a total privacy nightmare. Plus within just a few days, they had a security incident, and now a lot of people are questioning if it’s even safe to use.
Sure, it’s cheaper than ChatGPT, but the more you use it, the more it feels like a Temu knockoff instead of something from Amazon.
Head-to-Head: DeepSeek vs Gemini for Recruiters
It's clear that both of these AI models have the potential. But if you had to choose just one, which one would come out on top? Let’s break it down and see how they’re reasoning things out—I’ll show you how they’re approaching the tasks we’ve given them.
For this test, I'll use the same prompt Glen used in his video.
please write me a maximally inclusive Boolean Search String to find forklift operators
Each section will include a reasoning part and the results generated by AI, and I’m sharing both so you can easily compare. This is the first result I got from the AI, but if I run the prompt again, the results will change a bit—that’s just how LLMs work.
DeepSeek Reasoning
Okay, the user wants a maximally inclusive Boolean search string to find forklift operators. Let me start by brainstorming all possible terms related to forklift operators. First, the job title itself. "Forklift Operator" is the obvious one, but there might be variations. Maybe "Fork Lift Operator" with a space, or "Forklift Driver". Also, some might use "Powered Industrial Truck Operator" as that's the official term from OSHA.
Then, different types of forklifts. There are various classes, like Class I, II, III, etc., but not sure if people include those in their resumes. Maybe terms like "Counterbalance Forklift", "Reach Truck", "Order Picker", "Pallet Jack" (both manual and electric). Oh, "Cherry Picker" might be another term, though that's more for aerial lifts. Wait, but sometimes people might mix terms. Also "Tow Motor" or "Tow-Motor" is a brand but sometimes used generically.
Certifications are important too. OSHA certification, but maybe written as "OSHA Certified" or "Certified Forklift Operator". Some might have "Forklift License" or "Forklift Certification".
Abbreviations: FLT for Forklift Truck, maybe. Or "PIT" for Powered Industrial Truck. Also, "Material Handling" as a broader term that includes forklift operation.
Synonyms for operator: operator, driver, technician? Maybe not technician. "Handler" is a stretch. "Worker"? Maybe. So possible variations: "Forklift Operator", "Forklift Driver", "Forklift Handler", "Forklift Worker".
Also, considering misspellings or alternative spellings. For example, "fork lift" vs "forklift". So I should include both. Maybe use wildcards if allowed. In Boolean, sometimes you can use asterisks for truncation, but depends on the database. Wait, the user asked for a Boolean search string, so maybe using ORs and parentheses.
So putting it all together. Start with the main titles: "Forklift Operator" OR "Fork Lift Operator" OR "Forklift Driver". Then add the different types of equipment: "Reach Truck Operator" OR "Order Picker" OR "Pallet Jack Operator". Include the terms like "Counterbalance", "Tow Motor", "Cherry Picker" maybe. Then certifications: "OSHA Certified" OR "Forklift License". Also include "Powered Industrial Truck" terms. Maybe include related industries or tasks: "Warehouse" OR "Material Handling" to capture context where forklift is used, but that might be too broad. Wait, the user wants to find forklift operators, so maybe focusing on the job titles and certifications is better to avoid including irrelevant results.
Also, think about alternative job titles like "Warehouse Associate" who might have forklift experience. But including that could bring in non-operators. Hmm. The user wants maximal inclusivity, so maybe better to include broader terms even if some false positives. So maybe add "Warehouse Worker" OR "Material Handler" combined with forklift terms.
Wait, but how to structure it. Maybe split into sections: job titles, equipment types, certifications, and related roles. Use OR within each section and AND between sections? No, because that would require all sections to be present, which would narrow it down. But the user wants maximally inclusive, so probably OR all the terms together. But need to check for the different ways these terms can appear.
Alternatively, group synonyms with ORs and use parentheses to manage the logic. For example:
("Forklift Operator" OR "Fork Lift Operator" OR "Forklift Driver" OR ...) AND (certification terms OR equipment terms). But if we use AND, it might require both, which could reduce results. Since we want maximum inclusivity, perhaps OR everything related.
Wait, maybe the best approach is to list all possible variations connected with OR, using wildcards where applicable. For example, "forklift* operator*" would catch plurals and variations. But not sure if the system allows truncation. The user specified Boolean, so maybe just explicit terms.
Also, consider abbreviations: FLT Operator, PIT Operator. Maybe include those.
So compiling all these terms, the search string would look like:
("Forklift Operator" OR "Fork Lift Operator" OR "Forklift Driver" OR "Forklift Technician" OR "Powered Industrial Truck Operator" OR "PIT Operator" OR "Material Handling Equipment Operator" OR "Warehouse Equipment Operator" OR "Reach Truck Operator" OR "Order Picker Operator" OR "Pallet Jack Operator" OR "Electric Pallet Jack Operator" OR "Counterbalance Forklift Operator" OR "Tow Motor Operator" OR "Cherry Picker Operator" OR "Forklift Handler" OR "Forklift Worker" OR "Forklift Certified" OR "OSHA Certified Forklift" OR "Forklift License" OR "Forklift Certification" OR "Forklift Licence" OR "FLT Operator" OR "Forklift Truck Operator" OR "Forklift Operator Certified")
Additionally, maybe include misspellings like "Forklift Operater" but that might be too much. Alternatively, using wildcards if the platform supports them, like "forklift operat*" to capture operator, operators, operating, etc. But again, depends on the system.
Including both "Licence" and "License" for different spellings (UK vs US). Also, "Certified" and "Certification".
So the final string would be a long list of OR'ed terms, each in quotes to capture exact phrases, and accounting for different terminologies, certifications, equipment types, and common variations. That should make it as inclusive as possible.
DeepSeek Results
Boolean Search String:
("Forklift Operator" OR "Fork Lift Operator" OR "Forklift Driver" OR "Forklift Technician" OR "Powered Industrial Truck Operator" OR "PIT Operator" OR "Material Handling Equipment Operator" OR "Warehouse Equipment Operator" OR "Reach Truck Operator" OR "Order Picker Operator" OR "Pallet Jack Operator" OR "Electric Pallet Jack Operator" OR "Counterbalance Forklift Operator" OR "Tow Motor Operator" OR "Cherry Picker Operator" OR "Forklift Handler" OR "Forklift Worker" OR "Forklift Certified" OR "OSHA Certified Forklift" OR "Forklift License" OR "Forklift Certification" OR "Forklift Licence" OR "FLT Operator" OR "Forklift Truck Operator" OR "Forklift Operator Certified" OR "Class I Operator" OR "Class II Operator" OR "Class III Operator" OR "Class IV Operator" OR "Class V Operator" OR "Stock Picker" OR "Warehouse Associate" OR "Material Handler")
Key Inclusions:
Job Titles: Covers common variations (e.g., "Driver," "Technician," "Handler").
Equipment Types: Includes specific machines (e.g., "Reach Truck," "Order Picker," "Counterbalance").
Certifications: OSHA compliance, licenses, and certifications.
Spelling Variations: Accounts for UK/US spelling (e.g., "Licence" vs. "License").
Industry Jargon: Terms like "Tow Motor" (brand-specific), "FLT" (Forklift Truck), or "PIT" (Powered Industrial Truck).
Related Roles: Expands to broader roles that may include forklift duties (e.g., "Material Handler," "Warehouse Associate").
Use Cases:
Recruitment: Scanning resumes/job boards for candidates.
Research: Aggregating data on forklift operators in a database.
Compliance: Identifying OSHA-certified operators.
For even broader results, add wildcards (e.g., forklift*
to capture plurals) if your search platform supports them. Adjust terms based on regional or industry-specific vocabulary.
Google Gemini
How does Gemini compare when it comes to privacy? Well, Google, for all its flaws, has a much longer track record when it comes to data security. They've been handling vast amounts of user data for years, and they're subject to much stricter privacy regulations in many parts of the world, including GDPR in Europe.
Google has also implemented various measures to protect user data within the Gemini ecosystem. They use encryption to secure data both in transit and at rest, and they have strict access controls in place to limit who can see your data. They also provide users with tools to manage their privacy settings and control how their data is being used.
Now, it's important to acknowledge that no system is 100% foolproof. Data breaches can happen anywhere, even at the biggest and most reputable companies. However, Google's established reputation, its commitment to data security, and its compliance with various privacy regulations make Gemini a more trustworthy choice for recruiters, at least for now.
Google Gemini Reasoning
Here's a thinking process to arrive at the inclusive Boolean search string for forklift operators:
Understand the Goal: The request is to create a maximally inclusive Boolean search string. "Maximally inclusive" means we want to capture as many relevant results as possible, even if it means including some slightly less relevant ones that we can then filter out later. We want to avoid missing good candidates.
Identify Core Keywords and Concepts: What are the essential terms related to "forklift operator"?
Primary Job Title: "Forklift Operator" is the most obvious.
Alternative Job Titles/Synonyms: People might use variations like "Fork Truck Operator," "Lift Truck Operator," or even just "Warehouse Operator" (if forklift operation is a key part of that role).
Equipment/Machine Names: Instead of "forklift operator," someone might list the type of forklift they operate, like "Forklift," "Reach Truck," "Pallet Jack," "Order Picker," "Cherry Picker," "Turret Truck," "Side Loader," "Rough Terrain Forklift." (Initially, I might just think of "forklift" and "pallet jack," but then broaden to consider different types).
Licenses/Certifications: Often, forklift operators need certifications. Keywords like "Forklift Certified," "Certified Forklift Operator," "Forklift License," "License to Operate Forklift" are relevant. Abbreviations like "OSHA Forklift Certification" (though less common in titles, good to keep in mind for skills/experience sections).
Skills/Responsibilities (Implicit in Job Titles): Words associated with the actions of a forklift operator: "Material Handling," "Warehouse Operations," "Logistics," "Inventory Management," "Loading," "Unloading," "Shipping," "Receiving," "Stocking," "Order Fulfillment." These are more supporting terms, less likely to be the primary search terms, but can broaden the search if job titles are vague.
Consider Boolean Operators and Search Logic: To be inclusive, we need to use OR to combine synonyms and variations. We'll use AND sparingly to refine if necessary, but the goal is broad initially. Parentheses () are crucial to group terms correctly and ensure the logic works as intended.
Structure the Search String – Iteration 1 (Job Titles): Start with the core job titles and their variations, using OR:
"Forklift Operator" OR "Fork Truck Operator" OR "Lift Truck Operator"
Expand with Equipment Names – Iteration 2: Add the different forklift types. It's important to consider if "forklift" alone is enough, or if we should be more specific. Let's try including various types, just in case someone lists the equipment they operate instead of the title. Also, "Pallet Jack" is often considered a type of forklift or related material handling equipment, even if it doesn't always require the same certification.
"Forklift Operator" OR "Fork Truck Operator" OR "Lift Truck Operator" OR "Forklift" OR "Reach Truck" OR "Pallet Jack" OR "Order Picker" OR "Cherry Picker" OR "Turret Truck" OR "Side Loader" OR "Rough Terrain Forklift"
Self-Correction: This is getting long. Is just "Forklift" sufficient to capture most cases where someone mentions the equipment? Probably. Let's simplify a bit. Also, "Cherry Picker" and "Order Picker" are sometimes considered distinct equipment but in a warehouse context, they are often operated by similar roles. Let's keep them. "Turret Truck" and "Side Loader" are more specialized, perhaps less common in general job titles – let's keep them for inclusiveness but prioritize the more common terms. "Rough Terrain Forklift" might be for a more specialized outdoor role. Let's include it for breadth but remember it might bring in some less relevant results.
Revised Iteration 2 (Slightly Simplified):
"Forklift Operator" OR "Fork Truck Operator" OR "Lift Truck Operator" OR Forklift OR "Reach Truck" OR "Pallet Jack" OR "Order Picker" OR "Cherry Picker" OR "Turret Truck" OR "Side Loader" OR "Rough Terrain Forklift" (Removing quotes around single word "Forklift" is generally fine, and might catch more variations).
Incorporate Certifications/Licenses – Iteration 3: Add terms related to certification. Think about different phrasing.
"Forklift Operator" OR "Fork Truck Operator" OR "Lift Truck Operator" OR Forklift OR "Reach Truck" OR "Pallet Jack" OR "Order Picker" OR "Cherry Picker" OR "Turret Truck" OR "Side Loader" OR "Rough Terrain Forklift" OR "Forklift Certified" OR "Certified Forklift Operator" OR "Forklift License" OR "License to Operate Forklift"
Consider Location/Industry Context (Initial thought, but might complicate things): Should we include location? Or industry? For maximally inclusive, probably not initially. We want to cast a wide net. Location and industry can be added as filters after the initial broad search.
Skills/Responsibilities – Iteration 4 (More Supporting, Less Primary): Should we include "Material Handling," "Warehouse," etc.? These are too broad on their own. Searching just for "Warehouse" would give tons of irrelevant results. They're more useful for refining a search if the initial broad search is too noisy. For now, for maximal inclusivity, let's leave these out of the primary search string. They can be used in conjunction with other criteria later if needed (like an AND with location or industry).
Phrase Searching (Quotes): Use double quotes " around multi-word phrases like "Forklift Operator" to ensure those exact phrases are matched. This is important for accuracy.
Case Insensitivity: Boolean search is usually case-insensitive, but it's good to be aware of this. "Forklift Operator" and "forklift operator" should be treated the same.
Testing and Refinement (If possible): If you have access to a search platform (like LinkedIn Recruiter, Indeed Resume, etc.), it's always best to test the search string and see what kind of results you get. Adjust based on the results. If you're getting too many irrelevant results, you might need to refine (make it less inclusive). If you're getting too few, you need to broaden it further. For example, testing might reveal that "Rough Terrain Forklift" brings in a lot of construction equipment operators, and you might decide to remove it if your focus is purely warehouse roles. However, for maximal inclusivity as requested, keeping it is probably better and filtering later is the way to go.
Final Review: Look at the final string and ask:
Is it using OR effectively to combine synonyms? Yes.
Are key terms included? Yes (job titles, equipment, certifications).
Is it likely to be maximally inclusive? Yes, it casts a wide net.
Is it too broad? Potentially, but that's the point of "maximally inclusive." Filtering can be done after the initial search.
This iterative process, starting with core concepts and gradually expanding and refining the search string while considering Boolean operators and the "maximally inclusive" requirement, leads to the final recommended search string.
Google Gemini Results
Okay, here's a maximally inclusive Boolean search string to find forklift operators, designed to capture a wide range of relevant profiles. I'll break down why each part is included and how to use it effectively.
The Search String:
("Forklift Operator" OR "Fork Truck Operator" OR "Lift Truck Operator" OR "Powered Industrial Truck Operator" OR "Warehouse Operator" OR "Material Handler" OR "Logistics Associate" OR "Distribution Associate" OR "Shipping and Receiving Clerk" OR "Order Picker" OR "Reach Truck Operator" OR "Pallet Jack Operator" OR "Cherry Picker Operator" OR "Turret Truck Operator" OR "Side Loader Operator" OR "Rough Terrain Forklift Operator" OR "Telescopic Handler Operator" OR "Electric Pallet Jack Operator" OR "Walkie Stacker Operator") AND ("Forklift" OR "Fork Truck" OR "Lift Truck" OR "Powered Industrial Truck" OR "PIT" OR "Reach Truck" OR "Pallet Jack" OR "Order Picker" OR "Cherry Picker" OR "Turret Truck" OR "Side Loader" OR "Rough Terrain Forklift" OR "Telescopic Handler" OR "Electric Pallet Jack" OR "Walkie Stacker") AND ("Operate" OR "Operating" OR "Drives" OR "Driving" OR "Experience" OR "Certified" OR "Certification" OR "Licensed" OR "License" OR "Trained" OR "Training" OR "Proficient" OR "Skilled")
Explanation and Breakdown for Maximum Inclusivity:
Part 1: Job Titles (The First Parentheses Group)
"Forklift Operator": The most common and direct title.
"Fork Truck Operator": A common synonym.
"Lift Truck Operator": Another frequent alternative.
"Powered Industrial Truck Operator": The more formal, OSHA-compliant term. This is very important for compliance-focused searches.
"Warehouse Operator": Many forklift operators are titled this, especially if forklift work is part of their broader warehouse duties. This is a key inclusion for broader searches.
"Material Handler": Similar to Warehouse Operator, forklift operation is often a core skill for Material Handlers.
"Logistics Associate" / "Distribution Associate": These titles can sometimes encompass forklift operation, especially in entry-level roles within logistics or distribution centers.
"Shipping and Receiving Clerk": May operate forklifts for loading/unloading.
"Order Picker": While "Order Picker" is also a type of forklift, some people use it as a job title. Including it here captures those who might specifically list this title.
Specific Forklift Types as Titles:
"Reach Truck Operator"
"Pallet Jack Operator"
"Cherry Picker Operator"
"Turret Truck Operator"
"Side Loader Operator"
"Rough Terrain Forklift Operator"
"Telescopic Handler Operator"
"Electric Pallet Jack Operator"
"Walkie Stacker Operator"
Including these specific types as job titles is crucial because some people might list their job title as the specific equipment they operate.
Part 2: Equipment Keywords (The Second Parentheses Group)
This section lists the equipment names themselves. This is important because:
Someone might list "Experience operating Forklift" in their skills or summary, even if their job title isn't explicitly "Forklift Operator."
It captures profiles where the job title is broader (like "Warehouse Associate") but they mention forklift operation in their skills or responsibilities.
It covers variations in terminology (e.g., "Fork Truck" vs. "Forklift").
Includes abbreviations: "PIT" for "Powered Industrial Truck."
The list of equipment names is repeated from the job titles section for consistency and completeness.
Part 3: Action & Qualification Keywords (The Third Parentheses Group)
"Operate" / "Operating" / "Drives" / "Driving": These verbs indicate someone actively performs the task of operating a forklift.
"Experience": Captures profiles that mention "Forklift experience."
"Certified" / "Certification": Essential for roles requiring certification.
"Licensed" / "License": Similar to certified, depending on the region/industry.
"Trained" / "Training": May capture recent graduates or those who have received internal training.
"Proficient" / "Skilled": Indicates a level of competence.
Boolean Logic Explained:
OR: Within each parentheses group, OR ensures that the search will match if any of the terms in that group are found. This is what makes the search inclusive.
AND: AND between the parentheses groups means that the search will only return results that contain at least one term from each of the three groups. This helps to narrow down the results to be more relevant to forklift operation while still being broad.
How to Use This Search String:
Platform Compatibility: This string is designed to be compatible with most Applicant Tracking Systems (ATS), job boards (like Indeed, LinkedIn, Monster), and resume databases that support Boolean search.
Copy and Paste: Simply copy and paste the entire string into the search bar of your chosen platform.
Refinement (Optional but Recommended):
Location Filtering: After running the initial search, use location filters provided by the platform to narrow down by geographic area.
Experience Level: If you need a specific experience level, you might need to add additional keywords related to years of experience (e.g., "2+ years experience"). However, be careful as this can reduce inclusivity.
Industry/Company Type: You can also filter by industry or company type after the initial search to further refine results.
Exclusion Keywords (Use Sparingly): If you are getting too many irrelevant results (e.g., too many "Warehouse Operators" who don't actually operate forklifts), you could consider adding exclusion keywords using NOT. However, be very cautious with NOT as it can easily exclude good candidates. For example, NOT "supervisor" NOT "manager" might help if you are looking for individual operators and not leads, but it could also exclude qualified candidates who have some leadership experience but are still willing to operate forklifts.
Why This String is Maximally Inclusive:
Broad Terminology: It covers a wide range of job titles and equipment names, including common variations and industry-specific terms.
Action-Oriented Keywords: It includes verbs and qualification terms to capture people who actively do forklift work, not just those who mention forklifts in a tangential way.
Minimizes Exclusions: It avoids overly restrictive AND clauses and NOT clauses in the core search, focusing on bringing in a wide net of potential candidates.
Important Considerations:
"Maximally Inclusive" vs. "Perfectly Targeted": This string is designed for inclusivity. It might return some results that are slightly less relevant. You will likely need to review profiles to filter out truly irrelevant candidates. However, the goal is to avoid missing good candidates due to overly narrow search terms.
Platform Limitations: Some platforms might have limitations on the length or complexity of Boolean search strings. If you encounter issues, you might need to break down the search into smaller parts and combine the results.
Continuous Refinement: Boolean search is an iterative process. Run the search, review the results, and adjust the string as needed to optimize for your specific needs.
By using this comprehensive Boolean search string, you significantly increase your chances of finding a wide pool of qualified forklift operator candidates. Remember to always review the results and refine your search strategy as needed.
DeepSeek vs Gemini
So, where does that leave us? While it's clear that both DeepSeek and Gemini are powerful AI tools, DeepSeek's reasoning abilities are intersting, and it's definitely an AI to keep an eye on. However, the significant privacy concerns surrounding it make it a risky proposition for recruiters at this time.
For now, Gemini appears to be the more reliable, faster and secure option for those of us looking to leverage the power of AI in our work. Its seamless integration with the Google ecosystem, its user-friendly interface, its customization options, and Google's stronger stance on data privacy make it a safer bet for handling sensitive candidate information.
But here's an interesting takeaway from this whole comparison: watching how both DeepSeek and Gemini approach problem-solving can actually teach us a thing or two about how to improve our own search strategies.
By observing how these AI models deconstruct problems and identify relevant information, we can gain valuable insights into how to craft better boolean searches ourselves. It's like getting a peek behind the curtain of how these powerful AI "think."
If you're curious to learn more about how AI can help you master the art of boolean operators, and find those hidden gem candidates, I highly recommend checking out this article: Testing AI for Boolean Search Strings: Which is Best? It's definitely an interesting topic how AI can help us to be more creative in creating boolean operators.
So, what are your thoughts on DeepSeek and Gemini?
Unlocking the Power of Boolean
You've seen how AI like Gemini and DeepSeek can analyze information and identify potential candidates. But what if you could "think" like an AI yourself, crafting laser-focused searches that uncover exactly the talent you're looking for?
Behind this paywall, I’ll show you how to craft effective Boolean search strings. Think of it as your training ground for becoming a Boolean pro. I won’t just hand you the fish—I’ll teach you how to fish, giving you the skills to create super-targeted searches for any role, anywhere.