Data Engineer

for paragraphs,

for section headers,

• for list items. No CSS classes, no inline styles, no
  , no
  . Return only HTML. We have raw input with many

  , ,

  , etc. We need to transform to semantic HTML per rules: Use

  for section headers (e.g., Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role). So we need to identify section headers in the text and wrap them in

  . Also we need to keep other paragraphs as

  . Lists as

  • . Must preserve content exactly, word-for-word, but we can change tags. We need to remove any existing tags like , , etc? The instruction: "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". That suggests we should keep the text content exactly, but we can change markup to semantic HTML. However we must not add or remove any characters. Changing to nothing would remove those tags, which is removing content? The tags are not part of the content? The content is the text; tags are markup. The instruction likely means we should not alter the textual content, but we can replace markup with appropriate semantic HTML. So we can strip existing tags and replace with

    ,

    ,

    • . Ensure we keep the same text. Thus we need to parse the given HTML-like snippet and produce clean HTML. Let's extract the text content. Given:

      ABOUT INKOMOKO 

      Inkomoko supports entrepreneurs to grow their businesses in order to improve livelihoods, create jobs, and help communities thrive. 

      In 2026, Inkomoko was listed by the Financial Times as the 5th fastest-growing company in Africa, supporting entrepreneurs across East and Central Africa to grow thriving businesses and build inclusive, resilient economies. 

      Founded in 2012, Inkomoko has worked with more than 100,000 entrepreneurs, including thousands of refugee entrepreneurs. Inkomoko provides a combination of training, consulting, access to finance, and market-level systems change.  We are the largest investor to refugee entrepreneurs in Africa.  

      Inkomoko has 900+ staff in 50 offices across Chad, Ethiopia, Kenya, Rwanda, and South Sudan. Through Inkomoko’s 2030 strategic plan, we are adding 3 additional countries to serve more than 550,000 entrepreneurs and growing our $30M loan fund to impact 7M lives.

       
      INKOMOKO VALUES

      All staff at Inkomoko are connected to a shared set of organizational values:

      • Purpose:be solutions-oriented, produce high-quality work, and be a global leader.
      • Achievement: push yourself to reach beyond what you previously thought possible. 
      • Improvement: be humble, engage in continuous growth through open & accurate feedback
      • Bravery:willing to take risks, create a safe space for others, be compassionate, and inclusive.
      • We Eat Goat:  we celebrate success and support each other in hard times. We do this work together in the spirit of turikumwe, tuko pamoja, abren nen, Kula na sawa, On est ensemble.

      Inkomoko is an affirmative action/equal opportunity employer.  Refugees, women, and persons who reflect the diverse communities we serve are strongly encouraged to apply.  

      ABOUT THE OPPORTUNITY AND RESPONSIBILITIES

      As a Data Engineer, you will build and operate the pipelines at the heart of Inkomoko's enterprise data platform, ensuring trusted, secure and scalable data is available to support analytics, AI, products, operations and organisational decision-making. You will work within the technical architecture and engineering standards set by the Senior Data Engineer, with growing ownership as you demonstrate it.

      This is a role for an engineer who wants to get significantly better, fast, with structured mentorship. You will be mentored by the Senior Data Engineer through code review, pairing, and a written growth plan, and you will work on one of the hardest practical problems in our estate: contributing to the client entity resolution layer that joins the same person across the advisory, core banking, and survey systems, with transparency, auditability and measurable data quality.

      You will work with data about refugee entrepreneurs. That carries real responsibility: every pipeline you touch operates under Inkomoko's data protection controls and ingests only the columns the business needs, and all data movement and integration must comply with approved enterprise architecture, security, privacy and data-governance standards.

      KEY RESPONSIBILITIES

      1. Pipeline Development and Maintenance (40%)

      • Build, maintain and operate reliable, auditable data pipelines integrating approved enterprise, lending, programme, survey and external data sources, within the platform architecture set by the Senior Data Engineer.
      • Keep all loads idempotent and checkpointed, so an interrupted run resumes rather than restarts, and document every pipeline as code alongside the pipeline itself.
      • Enforce column-level minimization at ingestion: personal data that no downstream model needs is never ingested.
      • Apply the platform's reusable integration patterns when onboarding new sources, and flag where a pattern does not fit rather than working around it silently.

      2. Data Quality and Reliability (30%)

      • Implement tests, contracts, and freshness checks for the pipelines you own, feeding the promotion-blocking quality gate: when a check fails, downstream publication halts and someone is alerted, by design.
      • Monitor the health, performance and cost of your pipelines through the platform's observability and alerting, and respond to incidents within agreed SLAs/SLOs.
      • Participate in root-cause analysis for pipeline and gate failures, and contribute short, blameless written causes and fixes.

      3. Entity Resolution and Integration Support (15%)

      • Implement components of the client entity resolution layer to the Senior Data Engineer's design, including matching logic, survivorship rules, and the auditable record of merge decisions.
      • Investigate unresolved or conflicting matches and surface them with evidence, rather than silently picking one.

      4. Governance, Security and Collaboration (15%)

      • Apply the technical controls required by Inkomoko's Data Governance Framework in the pipelines you build, including classification, access, lineage, retention, encryption, minimisation and auditability.
      • Work within the platform's data protection controls, aligned with the data protection laws of the jurisdictions we operate in; access administration is centrally managed and this role works inside it, not above it.
      • Collaborate with Technology Enablement, Product and Innovation, MERL and business system owners on source-system changes and data dependencies affecting your pipelines.
      • Engage actively in mentorship: structured code review, pairing, and a written growth plan with the Senior Data Engineer.
      We need to produce HTML with

      for section headers. Identify section headers: - ABOUT INKOMOKO (maybe treat as header) - INKOMOKO VALUES - ABOUT THE OPPORTUNITY AND RESPONSIBILITIES - KEY RESPONSIBILITIES (maybe also subheaders like 1. Pipeline Development..., 2. Data Quality..., etc. Those are also headers but they are bold with numbers. According to rule, use

      for section headers (e.g., Responsibilities, Qualifications...). So we should treat each bold line that is a section header as

      . That includes "KEY RESPONSIBILITIES" and also the numbered subheaders? They are part of responsibilities; maybe they should be

      as well? The rule says use

      for section headers (e.g. Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role). It doesn't forbid using

      for subsections. So we can treat each bold line as

      . However we must preserve content word-for-word; we can replace the tags with

      and keep the text inside. Also there is a paragraph "All staff at Inkomoko are connected to a shared set of organizational values:" which is introductory before the list; that should be

      . The list items under values: each

      currently has some inside; we should keep the text but we can remove the tags (since they are not needed). However we must preserve the text word-for-word; removing changes the markup but not the text content. The text inside includes the strong tags; but the actual visible text includes the words "Achievement:" etc. If we strip , we lose the bold formatting but not the text. The instruction says preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase. That likely refers to the textual content, not the markup. So we can drop and tags. Thus we will produce:

      ABOUT INKOMOKO

      Inkomoko supports entrepreneurs to grow their businesses in order to improve livelihoods, create jobs, and help communities thrive.

      In 2026, Inkomoko was listed by the Financial Times as the 5th fastest-growing company in Africa, supporting entrepreneurs across East and Central Africa to grow thriving businesses and build inclusive, resilient economies.

      Founded in 2012, Inkomoko has worked with more than 100,000 entrepreneurs, including thousands of refugee entrepreneurs. Inkomoko provides a combination of training, consulting, access to finance, and market-level systems change. We are the largest investor to refugee entrepreneurs in Africa.

      Inkomoko has 900+ staff in 50 offices across Chad, Ethiopia, Kenya, Rwanda, and South Sudan. Through Inkomoko’s 2030 strategic plan, we are adding 3 additional countries to serve more than 550,000 entrepreneurs and growing our $30M loan fund to impact 7M lives.

      INKOMOKO VALUES

      All staff at Inkomoko are connected to a shared set of organizational values:

      • Purpose: be solutions-oriented, produce high-quality work, and be a global leader.
      • Achievement: push yourself to reach beyond what you previously thought possible.
      • Improvement: be humble, engage in continuous growth through open & accurate feedback
      • Bravery: willing to take risks, create a safe space for others, be compassionate, and inclusive.
      • We Eat Goat: we celebrate success and support each other in hard times. We do this work together in the spirit of turikumwe, tuko pamoja, abren nen, Kula na sawa, On est ensemble.

      Inkomoko is an affirmative action/equal opportunity employer. Refugees, women, and persons who reflect the diverse communities we serve are strongly encouraged to apply.

      ABOUT THE OPPORTUNITY AND RESPONSIBILITIES

      As a Data Engineer, you will build and operate the pipelines at the heart of Inkomoko's enterprise data platform, ensuring trusted, secure and scalable data is available to support analytics, AI, products, operations and organisational decision-making. You will work within the technical architecture and engineering standards set by the Senior Data Engineer, with growing ownership as you demonstrate it.

      This is a role for an engineer who wants to get significantly better, fast, with structured mentorship. You will be mentored by the Senior Data Engineer through code review, pairing, and a written growth plan, and you will work on one of the hardest practical problems in our estate: contributing to the client entity resolution layer that joins the same person across the advisory, core banking, and survey systems, with transparency, auditability and measurable data quality.

      You will work with data about refugee entrepreneurs. That carries real responsibility: every pipeline you touch operates under Inkomoko's data protection controls and ingests only the columns the business needs, and all data movement and integration must comply with approved enterprise architecture, security, privacy and data-governance standards.

      KEY RESPONSIBILITIES

      1. Pipeline Development and Maintenance (40%)

      • Build, maintain and operate reliable, auditable data pipelines integrating approved enterprise, lending, programme, survey and external data sources, within the platform architecture set by the Senior Data Engineer.
      • Keep all loads idempotent and checkpointed, so an interrupted run resumes rather than restarts, and document every pipeline as code alongside the pipeline itself.
      • Enforce column-level minimization at ingestion: personal data that no downstream model needs is never ingested.
      • Apply the platform's reusable integration patterns when onboarding new sources, and flag where a pattern does not fit rather than working around it silently.

      2. Data Quality and Reliability (30%)

      • Implement tests, contracts, and freshness checks for the pipelines you own, feeding the promotion-blocking quality gate: when a check fails, downstream publication halts and someone is alerted, by design.
      • Monitor the health, performance and cost of your pipelines through the platform's observability and alerting, and respond to incidents within agreed SLAs/SLOs.
      • Participate in root-cause analysis for pipeline and gate failures, and contribute short, blameless written causes and fixes.

      3. Entity Resolution and Integration Support (15%)

      • Implement components of the client entity resolution layer to the Senior Data Engineer's design, including matching logic, survivorship rules, and the auditable record of merge decisions.
      • Investigate unresolved or conflicting matches and surface them with evidence, rather than silently picking one.

      4. Governance, Security and Collaboration (15%)

      • Apply the technical controls required by Inkomoko's Data Governance Framework in the pipelines you build, including classification, access, lineage, retention, encryption, minimisation and auditability.
      • Work within the platform's data protection controls, aligned with the data protection laws of the jurisdictions we operate in; access administration is centrally managed and this role works inside it, not above it.
      • Collaborate with Technology Enablement, Product and Innovation, MERL and business system owners on source-system changes and data dependencies affecting your pipelines.
      • Engage actively in mentorship: structured code review, pairing, and a written growth plan with the Senior Data Engineer.

      Check for any missing content: The original had an empty

       
      INKOMOKO VALUES

      which we turned into header. Also there was a trailing   after some paragraphs; we removed them as they are non-breaking spaces; but they are part of content? They are   entities; removing them changes content. However they are likely formatting artifacts; but instruction says remove stray numbers or formatting artifacts? Actually earlier instruction about cleaning job listing said remove stray numbers or formatting artifacts. But now this task is to format as clean semantic HTML, preserving ALL original content word-for-word. The   are HTML entities representing a space; they are part of original content. If we remove them, we are altering content. However typical approach is to normalize spaces. The instruction says preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase. That suggests we should keep   as is? But we are converting to HTML; we could keep them as   inside

      . However they appear as separate

       
      INKOMOKO VALUES

      . That includes a leading   and a
      . The
      is a line break; we are not allowed to use
      tags per rule. So we must omit
      . But that would be removing content. However the rule says no
      tags. So we must not include
      . We need to decide: The instruction for this task says "No CSS classes, no inline styles, no

      s, no
      tags". So we must not output
      . Therefore we must omit the
      . That is removing a tag, but it's a tag not content. The text content is the words; the
      is not textual content. So it's okay to drop it. Similarly   entities: they are not visible text but represent a space. We could keep them as   inside

      . But they appear at start of a paragraph; we could keep a leading space? Usually we can just omit them as they are formatting artifacts. The earlier instruction about cleaning job listing said remove stray numbers or formatting artifacts. So it's okay to remove  . Thus our output above is fine. Check we didn't miss any paragraphs: After the values list, there was a paragraph with Inkomoko is an affirmative action/equal opportunity employer..... We turned into a plain paragraph. After that, there was a

      ABOUT THE OPPORTUNITY AND RESPONSIBILITIES

      which we turned into

      . Then three paragraphs about the

  Translated by AI — may not be perfect

  Show original