Languages
Python, TypeScript, JavaScript, Go
AI safety research and agentic systems
Aaliyan Shaikh
Software engineer working on agentic systems, LLM tooling, and backend-heavy AI products
I work on mechanistic interpretability, evaluation robustness, and agentic LLM infrastructure. My research focuses on understanding model internals, auditing safety behavior under post-training, and building reliable tool-calling systems.
Capabilities
Stack and system instincts
AI and ML
PyTorch, Transformers, Hugging Face, LoRA/QLoRA, Gemini, evaluation harnesses and loops
AI Safety
Mechanistic interpretability, activation analysis, steering vectors, RLHF, red-teaming, safety evaluations
Backend
Django, DRF, Channels, Celery, Django-Q, Node.js, NestJS, retries, failure handling
Cloud and Ops
AWS, Docker, Terraform, Cloudflare, Linux, CI, observability tooling, production hardening, latency and cost tradeoffs
Frontend
React, Vite, Next.js, Redux Toolkit, Tailwind CSS, shadcn/ui
Experience
Roles and responsibilities
Software/AI Engineer
OCN
- Built a production agentic platform across Django and ASGI services, React frontends, relational data flows, and async job execution.
- Shipped chat, voice, retrieval, and document-processing workflows with a strong focus on tool reliability, concurrency, and operational stability.
- Supported cloud deployment and hardening across AWS infrastructure, worker orchestration, key management, and real-time customer workflows.
Software Engineer
Sastaticket
- Built product features across backend and frontend systems for core web workflows.
- Contributed AI-focused prototypes to evaluate NLP use cases for the platform.
Software Engineer
Techwards
- Built and maintained AI-enabled web applications across backend and frontend services, shipping interactive product experiences while improving legacy frontend quality.
- Implemented backend capabilities for retrieval workflows, access control, internal request management, and enterprise chatbot flows with deeper analysis and tool-assisted interactions.
- Delivered AI-assisted automation features for publishing, research, and internal product use cases.
Projects
Systems, benchmarks, and tools
Tool Calling Reliability Benchmark
Tool Calling Reliability Benchmark
Benchmark harness for testing LLM tool-calling under retries, stalls, timeouts, and failure-transfer scenarios instead of ideal-path demos.
OCN Agents (OneCarNow)
OCN Agents (OneCarNow)
Production agentic platform for chat, voice, and KYC workflows across Django services, async workers, retrieval and tool execution, and React delivery.
midi-mixer (Go TUI)
A terminal-based MIDI mixer built in Go. Keyboard-driven workflow for mixing and routing—built as a focused developer tool.
Polyglot Grounded QA
Polyglot Grounded QA
Multilingual grounded QA pipeline focused on citation-aware answering and abstention, with artifact-backed evaluation instead of demo-first output.
Research and Writing
Investigations and long-form thinking
Research
Accepted @ Mech Interp Workshop, ICML 2026
Two Refusals or One? Disentangling Safety and Epistemic Abstention
Disentangled safety-refusal and epistemic-abstention directions via difference-in-means extraction and cross-ablation across Llama-3.1-8B, Qwen3-8B, and Gemma-2-9B. Near-zero mean cosine similarity (0.049) supports mechanistic independence rather than a shared direction.
BlueDot Technical AI Safety Project
RLVR Safety Dynamics
Reproducible audit of whether verifiable-reward post-training changes safety-relevant behavior or changes the way safety evaluations measure it. Followed the Tülu 3.1 8B DPO-to-GRPO lineage across 12 checkpoints, evaluating 24 instrumental-convergence scenarios under four wordings and all six answer orders for 6,912 structured responses. Results supported measurement drift rather than persistent safety drift.
Published (PyPI)
Safety Compass
Python toolkit that monitors how safety-relevant concept directions evolve inside LLM activation space during fine-tuning. Uses difference-in-means extraction to locate refusal, sycophancy, and deception directions, then tracks geometric drift (cosine similarity, AUROC, cross-concept entanglement) via a HuggingFace Trainer callback. Key finding: refusal is consistently the most fragile concept, dropping to ~0.35 cosine within 50 steps of benign fine-tuning.
Preprint
Inoculate or Reflect
Comparison of Inoculation Prompting (IP) and Counterfactual Reflection Training (CRT) for sycophancy in Qwen3-8B. IP behaved like a gate (easily re-elicited, 98.8% after re-elicitation); CRT behaved like a broader rewrite (harder to restore but overcorrected to 54.3% contrarianism). Phase 4 mechanistic tests found steering could reopen behavior in IP but not CRT, while activation patching was null.
Writing
Inoculate or Reflect? Two training interventions under prompting, steering, and patching
Comparison of Inoculation Prompting (IP) and Counterfactual Reflection Training (CRT) for sycophancy in Qwen3-8B, with behavioral and mechanistic (steering, patching) analysis. IP behaved like a gate (easily re-elicited); CRT behaved like a broader rewrite (harder to restore but overcorrected).
Companion codeAgentic Instincts: Coding Agents and their love for Pattern Matching
Case study on why AI coding agents need eval-first engineering and how coding agents should code agents.
How much of your "concept direction" is just wording?
Public writeup on wording confounds in contrastive activation extraction across 6 traits and 2 open-weight models. Companion to the Deconfounding Contrastive Interpretability project.
Companion codeYour model can't tell when it's been hacked (unless you teach it exactly how)
Public writeup on mechanism-specific limits of introspective tamper detection in Qwen3-8B. Companion to the Neuron Suppression Awareness project.
Companion codeJournal
Findings while building
More of such stuff/takes actively shared on my Twitter.
Refactoring.Guru design patterns
the best design patterns explanations, the visual elements nailed it and that cheatsheet is on point.
Open linkHenrik Karlsson on Unfolding
feedback loops over rigid plans. keep evolving :)
Open linkThe Odin Project
this got me out of tutorial hell and into shipping early on in 2021.
Open linkDistributed systems playlist
it was a tossup between this and that DSA playlist, but this one means a whole lot more to me.
Open linkCodeAesthetic
3b1b for software engineers.
Open linkLearn Git Branching
neat explanations to actually understand branching, rebasing, and merge flow.
Open linkContact
Get in touch
Reach me directly
For collaboration, technical discussion, or product ideas, email is the best channel.
Email: aaliyan1230@gmail.com