Skip to content
View 0xSuleman's full-sized avatar
1 follower · 1 following

Achievements

Achievement: YOLO

Achievements

Achievement: YOLO

Block or report 0xSuleman

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
0xSuleman/README.md

Suleman Ahmed

AI Engineer building reliable LLM workflows, reproducible ML systems, and data-quality-first automation

I am a software engineering student in Lahore, Pakistan, focused on AI systems that are measurable, explainable, and safe to operate. My work spans PyTorch research, schema-constrained LLM pipelines, deterministic extraction and QA, local-first AI, and full-stack product engineering.

Featured work

Project Engineering focus Portfolio context
Gap-Aware Bi-Mamba Leakage-conscious OULAD research, matched model comparisons, reproducible JSON artifacts Original research artifact
RALA Formula Lab Safe AST formula evaluation, rank diagnostics, matched baselines, reproducible result graphs Original research tooling; portfolio release published
TenderLo Deterministic extraction, evidence provenance, confidence-based QA, RLS End-to-end product engineering
ContentGenie Grounded generation, schema validation, editorial evaluation, media-rights manifests Downstream engineering based on ShortGPT; attribution documented
Menie Local-first transcription, diarization, summarization, privacy controls Experimental derivative of Meetily; attribution documented
Swift Hire CV parsing, ATS scoring, scheduling, audit logs, full-stack workflows FAST-NUCES team project; personal scope documented

What I care about

  • Reliable LLM applications: structured outputs, runtime validation, grounding, retries, and human review.
  • Reproducible ML: leakage-safe splits, validation-only selection, machine-readable results, and honest limitations.
  • Data quality: provenance, normalization, deduplication, confidence scoring, and auditable QA paths.
  • Local and deterministic AI where privacy, cost, or explainability makes hosted inference the wrong choice.
  • Shipping evidence: automated tests, CI, releases, demonstrations, and documentation that matches reality.

Core stack

Python · PyTorch · FastAPI · TypeScript · Next.js · React · Java · Spring Boot · PostgreSQL · Supabase · SQL · Docker · GitHub Actions

Contact

Pinned Loading

  1. gap-aware-mamba-student-risk gap-aware-mamba-student-risk Public

    Reproducible gap-aware state-space research for early student-risk prediction on OULAD

    Python

  2. rala-formula-lab rala-formula-lab Public

    Reproducible PyTorch workbench for RALA, hybrid linear attention, rank diagnostics, and evidence-first experiments

    Python

  3. contentgenie contentgenie Public

    Grounded and reviewable AI workflow for producing original YouTube Shorts, derived from ShortGPT

    Python

  4. Menie Menie Public

    Experimental local-first AI meeting assistant derived from Meetily

    TypeScript

  5. swift-hire swift-hire Public

    Team-built recruitment automation platform with CV parsing, ATS scoring, and scheduling

    JavaScript

  6. TenderLo TenderLo Public

    Evidence-backed Pakistan tender intelligence with deterministic extraction, QA, compliance, and scoring

    TypeScript