DeepSeek Mastery — Reasoning Models Without the Bill

Week 1 — meet the tool, get an account, and learn the screen before you learn the prompting.

1. Meet DeepSeek — what it is and who makes it

   DeepSeek is DeepSeek's reasoning-focused models served cheaply and published openly. It is at its best at showing its working on a problem with steps — arithmetic, logic, a small proof, a data question. It is at its weakest at being available and predictable: it is a smaller service than the frontier providers, and capacity shows — and knowing both halves is what separates somebody who uses it well from somebody who trusts it blindly.

   The lesson

   DeepSeek is a reasoning-focused models served cheaply and published openly made by DeepSeek. The models behind it are the DeepSeek chat and reasoning models, with open weights for the core ones. None of that matters on its own — what matters is that you know what kind of worker you have hired. At showing its working on a problem with steps — arithmetic, logic, a small proof, a data question is the job you hand it. At being available and predictable: it is a smaller service than the frontier providers, and capacity shows is the job you keep.

   Most people come to a course like this expecting a list of magic words. There is no such list. What there is, is a tool that produces an answer preceded by visible reasoning, which is either the most useful thing you will read today or a distraction, and produces it at a speed no human matches — which means the skill is not in writing the prompt, it is in knowing what a good answer looks like so you can tell the difference.

   Here is the first thing to try, worded the way you will word things for the rest of the course:

   PromptCopy

   A college has 1,240 students. 55% are in engineering, of which one fifth are in computer science. How many computer science students are there? Show every step, state your assumptions, and mark the answer clearly.

   Read the answer twice. The first read is for the content; the second is for the shape — did it answer the question you asked, or the question it found easiest? That second read is the habit this whole course is built on.

   Example A question you can verify in your head in ten seconds, answered with working you can inspect — the fastest way to see what a reasoning model is for.

   Practice Open DeepSeek, ask it the one question you would normally put to a search engine about your own field, and write down two things: whether the answer was right, and whether it would have taken you longer to find it yourself.

2. Signing in: what is free, what is paid, and what you actually need

   You do not need the paid plan to finish this course. Start on the free tier — the chat app is free to use, and the open weights can be downloaded and run yourself. Upgrade only when you hit a wall you can name: a longer file, a newer model, or a rate limit you keep meeting.

   The lesson

   Every one of these tools has a free tier that is good enough to learn on and a paid tier that removes a limit. the API is priced far below the frontier labs per token, which is why small teams build on it. The mistake is buying the paid plan in week one, before you know which limit you hit — you end up paying to remove ceilings you were never going to touch.

   Work out your own honest usage first. How many questions a day do you actually ask? How big are the files you upload? Do you need the newest model, or the fast one? For revision, for drafting, for coursework, the answer is usually the free tier.

   If your college or workplace provides an account, use it: self-hosted weights or a managed provider inside your own network is how most people in a job get access, and asking your placement cell whether one exists costs nothing.

   Example A half-hour reasoning pass on a maths or data problem costs nothing in the chat app, and the visible working is the part that teaches you.

   Practice Create the account, find the plan page, and write down in one line which limit you would hit first in your own week. It is usually a message cap or a file-size cap, not the model.

3. The screen: where every control lives

   A tour of the interface you will live in — a chat with a reasoning toggle, where the thinking appears above the answer and can usually be collapsed. Every panel has a reason to exist, and half of them are the difference between a chat and a system.

   The lesson

   The interface of DeepSeek is a chat with a reasoning toggle, where the thinking appears above the answer and can usually be collapsed. That sentence is worth slowing down on, because the single biggest cause of bad output is not a bad prompt — it is a good prompt typed into the wrong place.

   The history list is your memory of what worked. Name your conversations. The temporary or private mode is for anything you would not want in an account's history. The settings panel holds the personal instructions that apply to everything, which is where your context belongs rather than repeated at the top of every message.

   Do this once, properly: ask one arithmetic question with reasoning on and time both the wait and the answer, then ask the same with it off. It takes fifteen minutes and saves you those fifteen minutes every week after.

   Example The toggle is the whole product decision: fast answers for chatting, slow answers for problems. Using the wrong one wastes either your time or your money.

   Practice Spend fifteen minutes doing nothing but clicking. Open every panel, rename one conversation, and save one setting you will want again. Fluency with the screen is what stops you re-explaining yourself every session.

Week 2 — how DeepSeek reads text, the four-part prompt, your own work, and what to do when the answer is wrong.

1. How DeepSeek reads what you type

   Context, instructions and roles. A reasoning model spends tokens thinking before it answers, so the same question costs more of your allowance than it would on a chat model. Understanding what the model can see — and what it has already forgotten — explains almost every disappointing answer you will get.

   The lesson

   A reasoning model spends tokens thinking before it answers, so the same question costs more of your allowance than it would on a chat model. This is the machinery. A model does not remember your last conversation the way a person does; it is handed text and asked to continue it well. Everything you want it to know has to be in that text, in the same window.

   There is a difference between a system instruction — the standing context, set once — and a message, which is this request. Put your standing context in the standing place. "I am a final-year mechanical engineering student applying for data roles" belongs in your profile, not typed again at the start of every chat.

   And there is a hard limit. When a conversation gets long enough, the earliest turns fall out of view. Symptoms: it contradicts an instruction you gave ten messages ago, or forgets the file you uploaded. The fix is a new conversation with a written summary of where you got to, not a longer argument.

   Example A simple rewrite asked in reasoning mode takes twenty seconds and gives no better a result. This is the cheapest lesson in the course.

   Practice Ask the same question twice: once on its own, once after a short paragraph of setup about who you are and what you need. Compare the two answers and write down the difference. That gap is the thing you are learning to control.

2. The four-part prompt: role, task, context, format

   Almost every good prompt has four parts: who the model should be, what it must do, the facts it must use, and the shape of the answer you want. Leave out the fourth and you get an essay when you wanted a table.

   The lesson

   Write the four parts as four lines, in this order. ROLE: who it should answer as. TASK: the single thing you want done, as an instruction, not a wish. CONTEXT: the facts, pasted, not referred to. FORMAT: the exact shape of the answer — "a table with three columns", "five bullets, no more than twelve words each".

   FORMAT is the part everybody skips and the part that saves the most time. An answer you have to restructure by hand was not really an answer. Ask for the shape you are going to use: if it is going into a slide, ask for slide bullets; if it is going into a spreadsheet, ask for rows.

   Here is the same request done both ways — first as people usually write it, then in four parts:

   PromptCopy

   Weak:  tell me about data analyst jobs
   
   Strong:
   ROLE:    A hiring manager for entry-level analytics roles in India.
   TASK:    List what you screen for in a fresher's first 30 seconds.
   CONTEXT: I am a 2026 B.Com graduate with Excel, basic SQL and one
            dashboard project. No internship yet.
   FORMAT:  A table: skill | what a fresher shows | what most get wrong.
            Five rows maximum. No introduction.

   The second prompt is not longer because long is good. It is longer because it contains four things the model cannot guess, and the guess is where the useless answer came from.

   Example "Show your working, then give the final answer on its own line" is the request that gets the useful half of this model — the working is the teaching.

   Practice Take a task you did last week without AI and write the prompt in four labelled lines. Then run it, and rewrite only the part that failed — not the whole prompt.

3. DeepSeek for the work you actually have

   Assignments, revision, email, applications, meeting notes. A placement aptitude problem, a statistics question, a data-cleaning decision: it can walk through the steps where a fast model guesses. The test of a tool is whether it removes an hour from your week, not whether the demo looked clever.

   The lesson

   The highest-value use of DeepSeek for a student or a fresher is not writing essays. It is compression: turning a 40-page chapter into the six things you actually have to remember, turning a messy set of notes into a revision sheet, turning a job description into a list of what to prove.

   The second highest is structure: given a blank page, ask for three possible outlines and pick one. Being stuck is usually a problem of options, not of effort.

   Here is a prompt worth keeping verbatim — it is the one that turns a document into something you can study:

   PromptCopy

   Solve this past exam question. Show every step.
   
   [question]
   
   Then give me:
   1. The answer on its own line, labelled ANSWER.
   2. The step where people most often make a mistake, and why.
   3. A second, faster method if one exists.
   
   If the question is ambiguous, say so before solving anything.

   Notice the last line. Asking for the gaps is asking the tool to mark its own work, and it is the single most useful line you can add to a study prompt: the material it could not summarise is the material you have not understood yet.

   Example Give it a past exam question and ask for the working, the answer, and the two most common ways people get it wrong.

   Practice Pick the one task you repeat every week — the one that is boring rather than hard — and rebuild it in DeepSeek today. Time it. Then keep the prompt that worked, saved and named.

4. When the answer is wrong: iterate instead of restarting

   A bad answer is information. Its failure is overthinking a simple request, or wandering in the reasoning and arriving at a confident wrong answer anyway. Keep the conversation, name the fault, and correct one thing at a time — restarting from scratch throws away everything the model has already got right.

   The lesson

   There are four things that usually went wrong, and each has a different fix. It answered a different question — restate the TASK as one sentence. It made things up — supply the facts yourself and say "use only these". It wrote too much — ask for the length first. It sounded like a machine — ask it to rewrite for one specific reader and cut every third word.

   Say the fault out loud, in the message. "This is too long for a WhatsApp message and it sounds like a brochure." A model cannot fix a problem you have not named, and naming the problem is also how you find out what you actually wanted.

   Follow-ups that work: "shorter", "only the parts that are true for a fresher", "rewrite the second sentence three ways", "what would you have to check before I send this?". That last one is worth using before anything goes out with your name on it.

   Example If the reasoning is long and the answer is wrong, do not re-ask. Point at the step that broke: "you assumed the tax was included — redo only steps 4 and 5".

   Practice Take the worst answer you have received this week and correct it in three follow-up messages without retyping the original prompt. Notice how much faster it converges.

Week 3 — documents and tables, Visible reasoning — and the open weights behind it, the API, and one boring task automated.

1. Files, tables and long documents

   Uploading a PDF, a spreadsheet or a screenshot and asking questions about it: It reads documents and spreadsheets like the others, and it is unusually good at explaining the logic of a calculation in one. This is where these tools stop being a chat and start being work, and where the failure modes are worth knowing.

   The lesson

   It reads documents and spreadsheets like the others, and it is unusually good at explaining the logic of a calculation in one. The two failures to expect are the ones nobody warns you about: a long document is summarised as it is read, so a detail on page 60 can be missed; and a table with merged cells or a scanned page is likely to be read wrongly.

   So the working method is: ask narrow questions, and ask the tool to quote the line it is answering from. "From the attached file only, what does clause 4.2 let me do? Quote the sentence." If it cannot quote it, it did not read it.

   For a spreadsheet, ask for the formula and an explanation instead of the computed column — give it the column names and ask for the logic in words before the formula — then you can fix it yourself next month when the columns change.

   Example Upload a sheet with a broken total and ask which formula is wrong and why — a reasoning model that can see the sheet is genuinely useful for this.

   Practice Upload one real document you already own — a syllabus, a marksheet, a project report — and ask three questions whose answers you already know. When it gets one wrong, read the passage it quoted before you blame the file.

2. Visible reasoning — and the open weights behind it — DeepSeek's own feature

   Two things together. The reasoning is on screen, so you can check the step where an answer went wrong instead of guessing; and the core weights are published, so the same model can run on your own machine or your own server.

   The lesson

   Because the working is where the learning is. A bare answer tells you what to write; the working tells you how to think, and it makes a wrong answer diagnosable rather than mysterious. That is worth more for an exam or an interview than a marginally better final line.

   Turn reasoning on for anything with steps, and off for everything else. Read the working from the end backwards, hunting for the first step you disagree with. Correct that step in a follow-up and let it redo only from there.

   Here is the shape of it, in the form you will actually use:

   PromptCopy

   Reasoning: ON
   
   A shop sells two products. Product A sells for 249 rupees and makes a 24% profit.
   Product B sells for 399 rupees and makes a 12% profit. A customer buys 3 of A and
   2 of B.
   
   What is the total profit in rupees?
   
   Show every step. State any assumption you make. Put the final answer on its own
   line prefixed with ANSWER. If a step has two possible readings, solve both and
   say which is more likely and why.

   The mistake is using reasoning mode for everything. It costs time and tokens for no benefit on the other eighty per cent of your requests.

   Example A statistics question comes back with six numbered steps, one of which is the wrong distribution. You correct that step, not the question, and the second answer is right.

   Practice Take one past exam question you got wrong, run it with reasoning on, and find the exact step where your method and the model's differ. That difference is your gap.

3. The DeepSeek API and the open weights and your first script

   DeepSeek serves its models over an API that is compatible with the OpenAI request shape and priced far below the frontier labs, and publishes the weights of its core models. You do not need this to finish the course, and you do not need it for a fresher job either — but an afternoon here is what turns "I have used DeepSeek" into "I have built with it", which is a different sentence in an interview.

   The lesson

   DeepSeek serves its models over an API that is compatible with the OpenAI request shape and priced far below the frontier labs, and publishes the weights of its core models. The idea is simple: the same model you have been chatting with also answers a web request, so you can put it inside a script, a spreadsheet or a page. A key identifies you; a request sends the text; a response comes back as data.

   The reason to try it once, even if you never build anything: it makes the chat version less mysterious. You see that the whole conversation is text in and text out, that your instructions are literally lines of a request, and that "the model" is one parameter among several.

   A first call looks like this:

   PythonCopy

   from openai import OpenAI
   
   client = OpenAI(
       api_key="YOUR_DEEPSEEK_KEY",
       base_url="https://api.deepseek.com",
   )
   
   r = client.chat.completions.create(
       model="deepseek-reasoner",
       messages=[{"role": "user", "content": "A train covers 240 km in 3h20m. What is its speed in m/s? Show every step."}],
   )
   
   print(r.choices[0].message.content)

   The mistake is building a user-facing product on a smaller provider's capacity without handling the busy periods. Retry with a delay, and show the user that the job is queued rather than failed.

   Example A batch job that solves a folder of problems and records the working is cheap enough to actually run — which is the reason people build on this rather than on a premium model.

   Practice Get a key, run one request that works, and change one word in it to see the answer change. That is the whole of the first afternoon.

4. Automate one boring task with DeepSeek

   Automation is not about building a system. It is about doing one repetitive job the same way every time, in less time than last time, and being able to do it again next month.

   The lesson

   Pick the task by how often it happens, not by how impressive it would be. A weekly report you can half-generate beats a clever pipeline you build once and never open again.

   A weekly batch that works through a folder of practice questions and records the method rather than only the answer.. Whatever you choose, write the steps back out in plain English afterwards — "Step 1, open the sheet, Step 2, paste the names —" because the written steps are what you follow when the tool changes next quarter.

   And keep a copy of the prompt next to the task. A prompt that lives only in your chat history is a prompt you will rewrite from scratch in March.

   Example Twenty aptitude questions run overnight, with the working saved next to each — a revision file you keep returning to.

   Practice Name the task you repeat most often that involves typing, then write the prompt for it and run it three weeks in a row from the same saved place. Three runs is the point at which you know whether it is genuinely automated.

Week 4 — a finished portfolio project, the privacy rules, the limits, and the DeepSeek certification paths.

1. Build the portfolio project: a worked-solutions file for one exam

   Take one paper you are sitting — placement aptitude, statistics, a university unit — and build a file of thirty problems with full working, the common mistake, and a one-line summary of the method for each type. It is the thing you will talk about in the interview, so it has to be small enough to finish in a fortnight and concrete enough to show a person in one minute.

   The lesson

   Take one paper you are sitting — placement aptitude, statistics, a university unit — and build a file of thirty problems with full working, the common mistake, and a one-line summary of the method for each type. Finish it before you start the next one. A half-built idea shows nothing; a small finished thing shows that you can finish.

   Use the tool as a collaborator, not an author: ask it for a plan, a critique and a checklist, and write the work yourself. In the interview the questions will be about the decisions — why this, why not that — and only the work you did yourself has answers.

   Write one paragraph beside the project: what problem it solves, what you used, and what you would do differently next time. That paragraph is the interview.

   Example Finished, it is a revision document you actually use, plus a short note of the four method mistakes you personally kept making.

   Practice Solve ten of them from memory a week later and mark yourself against your own working. The ones you still get wrong are the chapters to reread.

2. Ethics, privacy and what never to paste

   These tools send what you type to somebody else's computer and keep it in a history. Never paste passwords, government ID numbers, bank or card details, medical records, or another person's private data — and never paste a company's confidential document.

   The lesson

   There is no version of this tool where your text stays on your laptop. Everything you type is sent to a server, kept in a history you can usually see, and may be reviewed or used to improve the product depending on the plan.

   For anything confidential, the correct answer on this course is the open weights: download the model and run it where the document already lives, rather than sending it to any provider at all.

   The practical rule for a fresher: replace the real thing with a stand-in. "Client A", "my friend's phone number", "the amount in the offer letter". The tool rarely needs the real value to do the work, and the stand-in costs you nothing.

   And the professional rule: an unpublished exam paper, a client's data or a colleague's private note never goes into a chat window, whichever provider it is. If your employer has an approved plan or a policy, that policy is the answer, not your judgement about how sensitive a file really is.

   Example A small, fast-growing provider is one you should read the data terms of before putting anything sensitive into it, and its terms are not the same as the frontier labs'

   Practice Go through your last five conversations and delete anything containing a real phone number, a client name, or a document you did not write. If you cannot find the delete button, that is the lesson.

3. Limits, hallucinations and how to check

   These tools predict plausible text. Long chains of reasoning look thorough and can be wrong at step three, with every later step built on the mistake That is a mechanism, not a moral failing — and the habit it demands is the habit of asking "where did this come from?" out loud, every time, before you use an answer.

   The lesson

   A model does not look things up unless it has been given a way to look things up, and even then it can attach a real number to the wrong claim. Long chains of reasoning look thorough and can be wrong at step three, with every later step built on the mistake

   So the rule is: numbers, dates, names, citations and legal or medical claims get checked in a primary source before they leave your hands. Everything else — drafts, structure, explanations, practice — is fair game.

   Three questions to ask before you trust an answer. Where did this come from? What would make it false? Who is the original source, and can I open it? If the third one has no answer, you have writing material, not facts.

   Verify the final answer with an independent method — a calculator, a spreadsheet, or your own hand-worked solution — never by asking the same model again.

   Example Give it a problem you can check in your head. The working will be tidy; the answer may not be. That is the whole lesson in ten seconds.

   Practice Ask for one statistic with its source, then open the source. Sometimes it exists. Sometimes the citation is invented, and the number is close enough to a real one to be dangerous. Either way you will remember the exercise.

4. Get certified: the official DeepSeek paths

   A Way2Fresher certificate for this course is free and lives on this site. Beyond it, DeepSeek publishes its own learning and certification material — and the rail on this course page links to it.

   The lesson

   There are two things called a certificate and they are not the same. The one this site issues records that you finished a structured course and built the project at the end of it — it is free, and it is yours to print. The ones DeepSeek issues record that you passed their own material.

   DeepSeek publishes its model cards, technical reports and API documentation; it does not run a certification programme. Its technical reports are unusually readable and worth an afternoon if you want to understand what a reasoning model actually does.

   Get both, in that order. The project is what an interviewer asks about; the certificate is what gets past a filter that looks for keywords. The links to the official paths are in the rail beside this lesson, each labelled with who issues it.

   And put the work on the certificate, not the other way round: a certificate with no project behind it is a line on a resume, and it lasts exactly until the first technical question.

   Example Solve one hard problem end to end and be able to explain every step in your own words. That ability, not the certificate, is what a technical interview measures.

   Practice Finish every lesson here, take your Way2Fresher certificate, then open one official path and work through it with the project you have already built. Being certified in the tool you can already use is a small additional step.