Skip to content
View DTiapan's full-sized avatar
🎯
Focusing
🎯
Focusing
5 followers · 10 following

Achievements

Achievement: Pull SharkAchievement: StarstruckAchievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: StarstruckAchievement: YOLOAchievement: Quickdraw

Block or report DTiapan

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
DTiapan/README.md

Ajas Bakran

AI Systems Engineer | AI Agent Evaluation & Reliability

I build and evaluate production-style AI systems, with a focus on AI agents, evaluation, reliability, and production engineering.

Focus Areas

AI Systems & Agent Engineering

  • Generative AI & LLM Applications
  • AI Agents & Agentic AI Systems
  • Agentic Workflows & Multi-Agent Systems
  • Agent Planning, Reasoning & Orchestration
  • Tool Calling & Function Calling
  • LLM APIs & Structured Outputs
  • Context Management & Context Engineering

AI Evaluation & Reliability

  • LLM / AI Agent Evaluation
  • Evaluation Frameworks & Evaluation Harnesses
  • Golden Datasets & Evaluation Datasets
  • Trajectory-based Evaluation
  • Tool-calling & Tool Correctness Evaluation
  • Planning, Reasoning & Task Completion Evaluation
  • LLM-as-a-Judge / Agent-as-a-Judge
  • Rubric-based Scoring & Judge Calibration
  • Regression Testing & Release Validation
  • Error Analysis & Failure Taxonomy
  • Production Readiness & Reliability

AI Safety & Responsible AI

  • Safety Evals
  • Red Teaming & Adversarial Testing
  • Prompt Injection & Jailbreaking
  • Robustness & Security Evaluation
  • Bias, Toxicity & Responsible AI
  • AI Safety & Alignment

RAG & Knowledge Systems

  • Retrieval-Augmented Generation (RAG)
  • Agentic RAG
  • Retrieval & Semantic Search
  • Embeddings & Vector Databases
  • Knowledge Retrieval & Grounded Responses
  • Data Ingestion & Transformation

Production AI Engineering

  • Python & FastAPI
  • Backend APIs & Microservices
  • PostgreSQL & Redis
  • Docker & Kubernetes
  • CI/CD & MLOps
  • Observability, Tracing & Logging
  • Monitoring, Scaling & Reliability

LLM Engineering

  • Prompt Engineering
  • Embeddings
  • Fine-tuning
  • Generative AI
  • Task-specific Reasoning
  • Dynamic Adaptation

Data & Evaluation Engineering

  • Python, Pandas & SQL
  • PyTest
  • Statistical Testing
  • Precision, Recall & F1
  • Calibration & Confidence Thresholds
  • Dataset Splits & Data Leakage
  • Hidden Test Sets & Benchmark Suites
  • Model Comparison & Regression Analysis

What I care about

Building AI systems that can be tested, measured, observed, and trusted — not just systems that produce impressive demos.

Writing & Content

Contact

Pinned Loading

  1. ai-agents-handbook ai-agents-handbook Public

    Welcome to AI Agents Handbook — your all-in-one guide for building AI agents, from basics to advanced techniques. Explore code, tutorials, and resources for beginners to experts!

    Python 86 18

  2. AgentEval AgentEval Public

    Open-source agent assurance: turn a PRD and agent URL into frozen test suites with evidence-backed PASS/FAIL/UNVERIFIABLE verdicts. Web console, REST API, SQLite. Domain packs for regulated teams (…

    Python

  3. Recall Recall Public

    Turnkey enterprise RAG: multi-format ingest, hybrid retrieval (Qdrant + BM25), reranking, citation-backed chat, workspaces, Firebase auth, golden-dataset eval gates, and GCP/Docker deploy—not a not…

    Python 1

  4. DocExtract DocExtract Public

    Production-grade multimodal document intelligence engine combining Vision Language Models (VLMs), spatial layout analysis, deterministic Pydantic mathematical cross-validation, and dynamic HITL rev…

    Python

  5. battery battery Public

    Local AI memory for coding agents: MCP server with hybrid BM25 + vector search, session capture, stale-memory prune, and git-committable BATTERY.md. SQLite + ONNX embeddings—no cloud DB or embed APIs.

    Python

  6. craft craft Public

    Pluggable orchestration for AI-assisted dev: phase router (using-craft), engineering ledger (DR/LL), and one-click install of curated upstream skills into .agents/skills/. IDE-agnostic—Cursor, Clau…

    Shell