AB

Reconcilliation Officer

Full time Head Office NBC, TZ
Posted 2 weeks, 4 days ago 47 views 0 applications

Job Description

Here's a thinking process: 1. **Analyze User Input:** - User is acting as a job listing editor. - I receive raw scraped text from a job website. - My job: extract ONLY actual job information, return clean, readable format. - Must REMOVE: social sharing text, author bylines, "you may also like", navigation, comments, ads/promotions, cookie/subscription notices, duplicate text, category labels not part of job, stray numbers/formatting artifacts. - Must KEEP and STRUCTURE: job title, company/organization, location, job description/responsibilities, requirements/qualifications, how to apply, salary (if mentioned), deadline/closing date, job ID/reference number, contract type/level. - Return ONLY the cleaned job text. No commentary, no headers like "Here is the cleaned version", no markdown formatting. Wait, the actual prompt I received is different. Let me re-read carefully. The user actually said: "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." Then there are rules: - Use

for introductory paragraphs and general text - Use

for section headers (e.g. Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role) - Use
  • for list items (duties, requirements, qualifications, skills) - No CSS classes, no inline styles, no
    s, no
    tags - Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase - If text has no clear sections, just wrap paragraphs in

    tags - Return ONLY the HTML. No markdown fences, no explanation, no preamble. Then there's a huge block of text that appears to be scraped job data. It's very messy with lots of nested ,

    , duplicates, etc. My task: Take that messy text, clean it according to the rules, and output clean semantic HTML using only

    ,

    ,
      ,
    • . No classes, no styles, no
      , no
      . Preserve all original content word-for-word. No adding/removing/summarizing. Let me look at the input text carefully. It's a job description for a reconciliation role at NBC (National Bank of Tanzania?). The input has: - Lots of outer wrapping ... lots of times - Some

      with "NBC is the oldest serving bank in Tanzania..." - Then

      empty paragraphs - Then

      Job Summary

      - Then description text about reconciliations - Then lots of stuff with job description header - Then
      • items with many duplicates - Then

        Policy, Procedure, Risk and Control Compliance — 30%

        - Then
        • items - Then

          Managing career and personal development —10%

          - Then
          • items (with some messy inline styles remnants like ;margin-top:0in;> but I need to preserve word-for-word? Wait, rule says "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". But also "No CSS classes, no inline styles, no
            s, no
            tags". There's a conflict. I need to interpret: I should remove inline styles but preserve the text content word-for-word. The rule "Preserve ALL original content word-for-word" likely means don't change the meaning/text, but I must remove inline styles as per the formatting rules. I'll strip style attributes but keep the text inside. - Then

            Role Specification:

            - Education requirement
            • ...> - Skills requirement
              • ...> - Then

                ... Qualifications... some text about qualifications with tags like "Analytical Thinking - Basic (Meets all of the requirements), Digital familiarity (Meets some of the requirements and would need further development), ..." This part is messy but I need to preserve word-for-word. - Then some trailing

                The challenge: The input has a lot of repetitive, nested, garbage structure. I need to extract the actual job content, remove the scaffolding (, empty

                , style attributes, duplicates), but keep all the actual job description text word-for-word. Let me parse and identify the actual meaningful content: - "NBC is the oldest serving bank in Tanzania with over five decades of experience. We offer a range of retail, business, corporate and investment banking, wealth management products and services." -> This seems like company intro, probably part of the job listing. - "Job Summary" section with text about reconciliations. - "Job Description​" header. - Then a big

                • list under "Daily Reconciliation, Investigation and Oversight — 60%". Many duplicates of the same li items. The rule says "Remove Duplicate text (if the title appears multiple times, keep it once)" but that was from the first system prompt. Wait, the user gave TWO prompts? Let me check. Actually, the user message starts with: "You are a job listing editor. You receive raw scraped text from a job website. Your job is to extract ONLY the actual job information and return it in a clean, readable format." Then there's a list of what to remove and what to keep. Then "Return ONLY the cleaned job text. Do not add any commentary, headers like "Here is the cleaned version", or markdown formatting. Just the clean job listing text." Then: "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." With different rules: use

                  ,

                  ,
                  • , no CSS classes/inline styles/
                    /
                    , preserve ALL original content word-for-word, if no clear sections just wrap paragraphs in

                    , return ONLY the HTML. No markdown fences, no explanation, no preamble. Then the actual job description text is provided. I need to follow the SECOND set of rules because it's the immediate task: format as clean semantic HTML with

                    ,

                    ,
                    • , no styles/classes/
                      /
                      , preserve all original content word-for-word. But the first set of rules also applies in spirit: remove social sharing, bylines, etc. The input doesn't seem to have those, but has lots of structural noise. Key instruction: "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". This is strong. I must not change any word. But I must remove inline styles, ,
                      ,
                      . How to reconcile? I'll strip the tags and style attributes, but keep the exact text content. The text inside the tags remains the same. I'll just restructure into

                      ,

                      ,
                      • . Let me go through the input text systematically, identify segments, and map them to HTML. Input text (I'll copy and clean mentally, but I need to output HTML): The raw text begins with: `... lots of nested spans ...

                        ` Then `

                        ` x2 Then `

                        Job Summary

                        ` Then `

                        ` x4 Then text: `To perform timely, accurate and complete reconciliations of assigned suspense, digital, Nostro, settlement and other general ledger accounts; identify, investigate, escalate and resolve reconciling items; maintain proper supporting documentation; and provide reliable reconciliation information to management to support financial control, regulatory compliance, loss prevention and effective decision-making.` Then `

                        ` x2 Then `... more spans ...Job Description​...` Then more spans with job description text? Actually after that, there's `
                        • Daily Reconciliation, Investigation and Oversight — 60%
                        ` and many
                      • items. Some are duplicated 3 times each. Then `

                         

                        ` then `

                        Policy, Procedure, Risk and Control Compliance — 30%

                        ` Then `
                        • ...
                        ` x? Then `

                        Managing career and personal development —10%

                        ` then `
                        • ...
                        ` Then `

                        Role Specification:

                        ` then `

                        Education requirement:

                        ` then `
                        • ...
                        ` Then `

                        Skills requirement:

                        ` then `
                        • ...
                        ` Then `

                        ...Qualifications......` and a list of qualifications/competencies: "Analytical Thinking - Basic (Meets all of the requirements), Digital familiarity (Meets some of the requirements and would need further development), Enabling team success (Meets some of the requirements and would need further development), Ethics and values (Meets all of the requirements), Financial Experience, Management Accounting (Meets some of the requirements and would need further development), National Diplomas and Advanced Certificates - Financial Sciences, Openness to change (Meets some of the requirements and would need further development), Quality orientation (Meets all of the requirements)" Then `

                        ` x2 Okay, now I need to output HTML using only

                        ,

                        ,
                          ,
                        • . No classes, no inline styles. I must preserve all original content word-for-word. So I will remove all , , ,
                          , style attributes, but keep the text. However, and are emphasis tags; the rules say "Use

                          for section headers... Use
                          • for list items". It doesn't say I can't use , but it says "No CSS classes, no inline styles, no
                            s, no
                            tags". It doesn't explicitly forbid or , but to be safe and semantic, I might use

                            for headers and not use unless it's part of the content. But the rule "Preserve ALL original content word-for-word" might mean I should keep the text as-is, including any bold markers? Actually, the content inside ... is text; I should keep that text, but can I keep the tags? The rule says "No CSS classes, no inline styles, no
                            s, no
                            tags". It doesn't mention or . But it says "Use

                            for section headers (e.g. Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role)". So I should use

                            for headers. I can probably replace Job Summary with

                            Job Summary

                            , but that would change the text? "Preserve ALL original content word-for-word" - if I change Job Summary to

                            Job Summary

                            , I'm changing the HTML structure but the visible text "Job Summary" remains. However, the rule might mean preserve the exact text content, not the tags. I think the intent is that the textual content should be preserved exactly, and I should format it using the allowed tags. I'll use

                            for section headers, and for any text that was inside or , I'll just keep the text, possibly without the tags, or I could wrap the header text in

                            and the text will be the same. But to be safe, I'll make the section headers as

                            followed by the header text, e.g., `

                            Job Summary

                            `. The original had `Job Summary`, the text is "Job Summary". I'll output `

                            Job Summary

                            `. Similarly for other headers. But wait: "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". This might mean I shouldn't even change the casing or wording of the headers. If I put it in

                            , the text "Job Summary" is still there. I think it's fine. Another issue: There are duplicate list items. The first system prompt said "Remove Duplicate text (if the title appears multiple times, keep it once)". But the second set of rules doesn't mention removing duplicates. It says "Preserve ALL original content word-for-word". So I should keep all duplicates? That seems odd for a job listing, but the rule is strict. I'll keep all text as appears, including duplicates. However, the duplicate li items are exactly the same text repeated 3 times. I'll keep all three repetitions as they appear in the source, because removing them would be removing content. But the rule "Preserve ALL original content word-for-word" means I shouldn't omit any part of the original text. So I'll include all duplicates. Wait, but the first prompt was from the user's system instruction, and the second is the actual task. The user might have combined two prompts, but I should follow the immediate instructions: "Your ONLY job is to format it as clean semantic HTML." with the rules listed. The rules include "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". So I must not remove duplicates. I'll keep them. However, there's also: "If text has no clear sections, just wrap paragraphs in

                            tags". The text has clear sections: Job Summary, Daily Reconciliation..., Policy..., Managing career..., Role Specification, Education, Skills, Qualifications. I'll use

                            for section headers. Let me plan the HTML structure: - Start with the company intro paragraph? The very first text: "NBC is the oldest serving bank in Tanzania with over five decades of experience. We offer a range of retail, business, corporate and investment banking, wealth management products and services." This is inside

                            initially, then lots of spans. I'll make it a

                            . - Then maybe a

                            Job Summary

                            - Then the reconciliation description paragraph. - Then

                            Daily Reconciliation, Investigation and Oversight — 60%

                            or just

                            Daily Reconciliation, Investigation and Oversight

                            ? The original had `Daily Reconciliation, Investigation and Oversight — 60%` as a list header before the
                            • . I'll use

                              Daily Reconciliation, Investigation and Oversight — 60%

                              - Then the
                              • items (with duplicates) - Then

                                Policy, Procedure, Risk and Control Compliance — 30%

                                - Then
                                • items - Then

                                  Managing career and personal development —10%

                                  - Then
                                  • items - Then

                                    Role Specification

                                    - Then `

                                    Education requirement:

                                    ` then `
                                    • ...
                                    ` - Then `

                                    Skills requirement:

                                    ` then `
                                    • ...
                                    ` - Then qualifications section: the text "Analytical Thinking - Basic (Meets all of the requirements), Digital familiarity (Meets some of the requirements and would need further development), Enabling team success (Meets some of the requirements and would need further development), Ethics and values (Meets all of the requirements), Financial Experience, Management Accounting (Meets some of the requirements and would need further development), National Diplomas and Advanced Certificates - Financial Sciences, Openness to change (Meets some of the requirements and would need further development), Quality orientation (Meets all of the requirements)" This was under `

                                    ...Qualifications...

                                    `. I'll make it a

                                    or maybe

                                    Qualifications

                                    followed by a

                                    with that text. The original had "Qualifications" as a heading inside spans. I'll use

                                    Qualifications

                                    then a

                                    with the competency list. - Then maybe closing paragraphs. But I need to be extremely careful to preserve all content word-for-word. Let me extract each segment's exact text. Let me re-read the input carefully, segment by segment, and note the exact text I must include. I'll copy the input text into analysis, but since I'm an AI, I'll parse it mentally. Input segments (cleaning tags but keeping text): 1. Company intro: "NBC is the oldest serving bank in Tanzania with over five decades of experience. We offer a range of retail, business, corporate and investment banking, wealth management products and services." 2. Empty paragraphs: `

                                    ` x2 and x4 - these are empty, I can ignore them or include empty

                                    ? The rule says "Preserve ALL original content word-for-word". Empty tags have no content, so I can omit them or include

                                    . I'll probably omit truly empty paragraphs to keep HTML clean, but the rule says don't remove content. I'll include them as

                                    maybe, but it says "No ... no
                                    tags". Empty

                                    is allowed? I'll see. Actually, the rule "If text has no clear sections, just wrap paragraphs in

                                    tags" suggests I should wrap text in

                                    . Empty paragraphs might be considered

                      Apply Now ↗

                      How well do you match?

                      Get an instant AI match score for this role — free, takes 3 minutes.

                      Tailor your CV for this role

                      The concierge rewrites your whole CV and writes a matching cover letter for this job — opens right here, nothing to paste.

                      Tailor My CV to This Job ✍️

                      Free cover letter for this job

                      Upload your CV and get a tailored cover letter in seconds — free, no account needed.

                      Generate a Cover Letter 📝

                      Join Our Tanzania Channels

                      Get free job alerts on your phone

MJC
ECHO
Your MJC Assistant

I'm ECHO, your MJC career assistant. I can help you find jobs, explore career tools, and connect with opportunities across Africa.

How was your experience with ECHO?