I sit between stakeholders and systems, uncover what the business truly needs, and build autonomous AI architectures that deliver it at scale. Over 10 years across BFSI, CPG, energy, and healthcare; MBA from Liverpool Business School, UK.
The 10-day process that becomes 2 hours. The 3% error rate that drops to 0.5%. The best AI systems are the ones nobody notices they just work.
Python |
Go |
Rust |
SQL |
LangChain |
LangGraph |
CrewAI |
smolagents |
LlamaIndex |
Pydantic AI |
Haystack |
Agno |
Mem0 |
MCP |
Dify |
n8n |
cognee |
Neo4j |
Ray |
OpenAI |
Anthropic |
Claude |
Gemini |
xAI |
DeepSeek |
Mistral |
Hugging Face |
Groq |
Ollama |
llama.cpp |
vLLM |
Perplexity |
Chroma |
Qdrant |
Weaviate |
Pinecone |
Milvus |
pgvector |
FAISS |
Spark |
Airflow |
MLflow |
W&B |
OpenTelemetry |
Prometheus |
Grafana |
AWS |
GCP |
Azure |
Docker |
Kubernetes |
Terraform |
GitHub Actions |
Streamlit |
Gradio |
FastAPI |
Redis |
Celery |
uv |
Cursor |
Windsurf |
Copilot |
Linux |
Git |
I fix real bugs in production-grade AI infrastructure agentic frameworks, LLM platforms, vector databases, AI tooling. The kind of issues that silently corrupt data, break under concurrency, or surface only at scale.
Merged
Notable open PRs
| Repo | Contribution |
|---|---|
| livekit/sip | In-dialog request routing via the INVITE's Record-Route (RFC 3261 §12.1.1); media-drain before BYE to prevent voicemail clipping |
| livekit/livekit | Room re-create race: pre-clear room state so same-name recreate gets a fresh SID |
| livekit/protocol | SIP header value cap raised to 4KB for carrier tokens (X-Twilio-CallToken) |
| NousResearch/hermes-agent | 11 open correctness PRs: progress-narration loop retry, RFC 2047-safe sender parsing, 4xx/5xx stream-error classification, digit-string list counters |
| langgenius/dify | PG logical replication fix; agent file type filtering; MCP client timeout fallback |
| chroma-core/chroma | Embedding-function config correctness, NUL-byte FTS5 corruption guard, client path normalization |
| crewAIInc/crewAI | Async native tool calling on the kickoff caller's event loop (#8354) |
| mem0ai/mem0 | Log redaction for sensitive content; embedding dims auto-propagation |
| huggingface/smolagents | Code agent reliability and tool-use edge cases |
| Blaizzy/mlx-audio | ModelProvider LRU eviction + idle TTL sweeper — fixes the GPU/MPS memory leak |
| andrewyng/openworker | Scheduled-run tool-role 400 + silent last_status=ok failure masking |
Open to architecture challenges, enterprise AI strategy, freelance engagements and projects, or startup ideas in the GenAI space. If you're building something that matters, let's talk.