Three tracks we build & share:
We skip the 101 slide decks. Everything here is open-source, reproducible, and written by engineers for engineers. Clone the starter repos, run the notebooks, and join #help on Discord if you get stuck.
Hack Nights: Live Code in the Room
Every session starts with a verified GitHub repository. We clone, configure dependencies, understand the math and memory constraints, and run inference or fine-tuning runs before leaving the room.
Local LoRA Fine-Tuning
Quantize base 8B models to 4-bit NF4, apply rank-16 LoRA adapters, fine-tune on custom technical markdown in 35 minutes, and export merged GGUF weights.
Production vLLM Inference Engine
Deploy continuous batching, chunked prefill, and OpenAI-compatible streaming endpoints on Linux instances with automatic GPU memory paging.
Hybrid RAG: BM25 + Dense Vectors
Construct a two-stage retrieval engine combining lexical keyword search with dense embeddings and reciprocal rank fusion to eliminate semantic hallucination.
Autonomous Agent Tool-Use Runtime
Build deterministic multi-turn tool loops with typed schema validation, automatic retry on truncated brackets, and human-in-the-loop operational gates.
Local Edge AI with ONNX & WebGPU
Run quantized small transformers directly inside client browser tabs and desktop apps with zero backend server roundtrips.
Synthetic Dataset Bootstrapping
Generate high-quality domain-specific instruction datasets using frontier models with automated deduplication and semantic contamination checks.
War Stories: What Failed at 2 AM
Unfiltered retrospectives on production crashes. We inspect real trace telemetry from silent vector drift, runaway API billing loops, and p99 inference spikes, breaking down exactly how to engineer resilient guardrails.
The $14,000 Infinite Recursive Tool Loop
Detailed post-mortem on how an unchecked agent spawned 4,200 nested sub-queries in 38 minutes, and the recursion depth tokens that permanently fixed it.
Chunking Size Disasters in Financial Docs
Why fixed 512-token windows sliced multi-column balance sheets in half and generated 40% error rates in financial summaries.
Cold-Start Inference Spikes on Serverless
Diagnosing 28-second cold starts with 7B models on serverless containers and migrating to persistent pre-warmed GPU pools.
Vector Drift After Embedding Updates
An accidental model version bump that silently degraded retrieval recall to 41% without triggering a single HTTP 500 error.
JSON Output Truncation on Edge Cases
Hard token limit crashes that severed closing braces, and why grammar-based constrained decoding saved production pipelines.
Memory Leaks in Async Model Handlers
CUDA memory fragmentation in Python async workers that forced daily container restarts and the precise torch.cuda.empty_cache() fix.
Paper Sprints: Research into Working Code
We read preprints the week they drop and build minimal, clean Python and CUDA proof-of-concepts to test whether novel architectures deliver their promised throughput and latency gains.
Speculative Decoding with Draft Models
Pair a compact 1B draft model with an 8B target model to verify multiple candidate tokens per forward pass and slash latency.
Mamba & State Space Architectures
Analyze selective state spaces, compare memory footings against quadratic attention, and benchmark throughput across 64K context tokens.
Test-Time Compute & Search Trees
Build multi-path candidate generation with reward verification models to solve complex algorithmic tasks during inference.
Vision-Audio Cross-Modal Projection
Pass raw spectrograms and visual patches directly to unified decoder transformer heads without converting to intermediate text.
Direct Preference Optimization (DPO)
Align models with pairwise preference loss directly on chosen vs rejected answers without running unstable PPO reinforcement loops.
Sparse Mixture of Experts (MoE) Routing
Implement auxiliary load-balancing loss, route tokens through top-k feedforward experts, and measure compute efficiency.
Want to build these live?
Grab your laptop and join fellow developers for Hack Night #04.
View Meetup Schedule →Have questions on an architecture?
Join our Discord community and discuss code with 2,500+ friendly builders.
Meet the Community →